Innovative technology advances TMP energy savings
Today's TMP mills are under constant pressure to reduce operating costs and improve pulp quality without capital expenditures. The TMP operation used as a reference in this paper is no exception: one of the primary objectives for 2002 was to save energy without any adverse effects on pulp quality.
The mill employed an innovative new technology in the Sunds RGP 76CD rejects refiners to achieve the energy saving goal – SmartPlates, which is one way of improving the refiner operation, through use of an array of temperature sensors embedded in the refiner plates. This system was employed with two goals in mind: (1) optimized refiner plate patterns and (2) improved refiner control.
Traditional methods of refiner plate optimization involve numerous trial and error iterations, which can be a lengthy, resource-consuming process. By using sensors to measure what is happening within the refiner as it is happening, we can gain a much better understanding of the process and the plate design impacts on refiner operation and pulp quality. The SmartPlates temperature measurements in combination with process information from the DCS allow us to understand where the energy is being applied and where the inefficiencies are within the current refiner plate design so that new designs can be developed that provide equal or better pulp quality at lower specific energy levels. This paper will explain the optimization process and show the results achieved with the new plate patterns.
In addition to plate pattern optimization, SmartPlates is also being used for closed loop control of the rejects refiners at the reference mill. The temperature profile within the refiner is used to control the dry fiber mass flow through the refiner and the refining consistency, leading to improved pulp quality stability. This paper will provide an explanation of the theory behind this control strategy and the results of running in control at this mill.
The combined results of plate optimization and refiner control have led to significant energy savings in the rejects refiners. Running in control has led to reductions in variability of blow-line consistency and reductions in variability of rejects latency chest freeness and fiber length. The results will be presented in more detail in this paper.
In recent years much of the North American pulp and paper industry has struggled with poor market conditions and rising costs, particularly the cost of electrical energy. For a mill to remain competitive it must produce a consistently high quality product that meets its customers' needs at the lowest possible cost.
In a typical TMP plant, the cost of electrical energy accounts for one third to one half of the total TMP manufacturing cost, so even a small reduction in energy consumption can lead to significant annual savings. Because of this, mills are constantly seeking new and innovative ways to save energy. The problem is that many solutions that offer energy savings do so at the expense of pulp quality. The key is to save energy without any adverse effects on pulp quality.
In addition to maintaining pulp quality, there has been much emphasis on reducing the variability of the pulp quality. A consistent product from the pulp mill will contribute to a consistent operation of the paper machine, which means fewer breaks and fewer defects in the sheet, and ultimately improved pressroom runnability. Pulp quality variability begins with the incoming raw material to the TMP mill. In many situations mills are now accepting chips from a wider variety of sources than once was common, leading to more variability in the incoming raw materials. It is important to find a way to compensate for this variability in raw materials in the TMP process so that the final pulp quality is stable.
The reference mill discussed in this paper is a newsprint manufacturer in eastern Canada, utilizing TMP, groundwood, deinked pulp, and occasionally kraft as the paper machine furnish.
The TMP mill has two primary, two secondary, and two rejects Sunds RGP 76CD pressurized refiners. There is a single stage of main line screens running wedgewire screen cylinders. There is also a single stage of rejects screens running wedgewire screen cylinders, followed by four stages of rejects cleaners.
SmartPlates is a refiner optimization and control system that utilizes temperature sensors embedded in an array in the refiner stator plates to provide information about what is happening inside the refiner in real-time. This system allows us to (1) optimize the plate patterns and the process conditions and (2) control the fiber mass flow and refining consistency. Previous studies have shown that refining zone temperature measurements are an effective tool for controlling a TMP refiner , , .
A previous paper by Johansson  provides a thorough explanation of the SmartPlates theory.
To summarize, the SmartPlates system measures the refining conditions within the refining zone in terms of temperature. It can be shown that most of the applied energy goes to a change in internal energy, Figure 1, and by applying the first law of thermodynamics, the temperature measurements can be used to evaluate the refining efficiency (as a function of plate pattern, position, and operating conditions). The software then computes process conditions such as refining consistency, applied power, or refining intensity as a function of radius. These computations are all based on fundamental laws of thermodynamics, and thus no "tuning parameters" are used. This information reveals the efficiency of the refining operation as a function of position, and can thus be used for plate optimization.
Most TMP refiners are fed volumetrically so when there are changes in the bulk density of the incoming raw material, the mass flow will change, therefore the refining action on the fibers is not constant, leading to pulp quality variability. With SmartPlates, the volumetric feed rate to the refiner (i.e. metering screw speed) is varied to keep the mass flow to the refiner constant by controlling the refining zone temperature. In addition, the dilution water flow to the refiner is varied to keep the refining consistency constant. This leads to constant refining action on the fibers and much less process variability.
SmartPlates can also be used to optimize the refiner plate patterns. The SmartPlates temperature measurements in combination with process information from the mill's DCS allow us to understand where and how the energy is being applied and where the inefficiencies are within the current refiner plate design so that new designs can be developed that provide equal or better pulp quality at lower specific energy levels.
The system comprises three main components: (1) the refiner plates themselves, containing the temperature sensors, (2) the control software, and (3) the control hardware, which includes the components required for the SmartPlates software to communicate with the mill's operating systems and equipment.
Plate Pattern Optimization
A structured approach was used to optimize the refiner plate patterns at the reference mill. First, a set of equivalent refiner plates, instrumented with SmartPlates sensors, was run in one of the rejects refiners. Tests were run in which operating parameters were varied according to a statistically determined design of experiments. The operating parameters investigated were: throughput, flat zone gap, CD zone gap, infeed dilution, CD dilution, infeed pressure, casing pressure. Pulp samples were taken from the blowline at each different set of operating conditions and tested.
The pulp quality, operating conditions, and temperature information were analyzed through advanced software to determine where and how the energy was being applied with this particular plate pattern combination. This software was further used to develop new plate patterns that would reduce the inefficiencies within the current plate designs to save energy without adversely affecting pulp quality.
At the reference mill, SmartPlates is used in the rejects refiners to control the dry fiber mass flow and pulp consistency through the refiner. Temperature and mass flow are controlled by manipulating the metering screw speed. Consistency is controlled by manipulating the infeed and CD dilution water flowrates.
RESULTS AND DISCUSSION
Plate Pattern Optimization
The first application of SmartPlates at the reference mill was to optimize the plate patterns used in the rejects refiners, to provide energy savings with no deterioration in pulp quality.
We began by running a set of plates that was quite similar to the previous standard rejects plates so that we could learn where and how the energy was being applied within the refining zone. Analysis of this information allowed us to apply a new plate design combination that would save energy.
Figure 2 shows the refining zone temperature profiles of the previous and new standard plates. The first five points of each profile correspond to the flat zone and the next five points are located in the CD zone. The new refiners plates have shifted the temperature peak from the flat zone to the CD zone.
The first set of optimized refiner plates were installed in R2 rejects refiner in late February 2002. The results were promising, so it was decided to proceed with the same plates in R1 rejects refiner in May 2002. Due to continuing positive results in terms of energy savings and pulp quality, these plates have continued to run in both rejects refiners since that time. Table 1 provides a summary comparing the performance of the previous standard rejects refiner plates with the new optimized plates, for all of 2002. The quality data shown in Table 1 is from the PQM, which samples after the rejects latency chest. Because the pulp from both rejects refiners goes to this common chest, the data shown is for one rejects refiner operation at a time, which is the normal operating situation for the mill.
The optimized rejects refiner plates have led to 9-11% (78-90 kWh/T) energy savings with improved pulp quality for both rejects refiners. For the same freeness, the fiber length is higher and there is a greater amount of long fibers with the optimized refiner plates. In addition, shive content is reduced and the fibers are less coarse. The improved pulp quality arises because more of the energy is now being applied in the CD zone, where the bars are finer and the refining action on the fibers is gentler – i.e. lower intensity. The energy savings are a result of the improved efficiency of transportation of the fibers and steam out of the flat zone plates and into the CD zone.
Once the optimal plate pattern was developed, the next application of SmartPlates at the reference mill was to provide closed loop control of the rejects refiner in order to reduce pulp quality variability.
The mill utilizes the PQM to monitor pulp quality from the rejects latency chest. When running in SmartPlates control , the variability of the PQM quality parameters is lower than when running in manual, as desired. Table 2 summarizes the results for 2002 for normal operation (i.e. greater than one-third of standard main line production rate) with the optimized refiner plates. The data shown in Table 2 is the standard deviation for each quality variable.
It is interesting to note that the reduction in standard deviation is greater for R1 than for R2, despite the fact that R1 produces pulp with less variability than R2 in manual.
The reductions in variability shown in Table 2 may not seem so dramatic at first glance, but upon consideration of the mill's actual operation, the results are quite significant. These considerations are explained in the following paragraphs.
First of all, the refined rejects chest has a residence time of about 30 minutes and the pulp is well-mixed before being tested by the Inline PQM. The SmartPlates controllers reduce variability in the 10-60 minute timeframe, which means that some of the pulp quality variability reductions at the refiner are not captured by the PQM due to the mixing action in the chest. If samples were taken at the blowline of each refiner and then tested, the reductions in standard deviations would be much greater than shown in Table 2.
Secondly, the TMP operators make frequent – i.e. a few times per shift – changes to the rejects refiner throughput to maintain the level in the unrefined rejects chest between certain minimum and maximum limits. The level in this chest depends greatly on the production rates through the main line refiners and the rejects rate from the screens. Each time the chest level becomes too low or too high, the operator must temporarily place the SmartPlates controllers in manual, make the production rate change, then re-engage the controllers. Because SmartPlates is meant to control the real tonnage rate through the refiner to reduce quality variability, anytime the operator manually changes the throughput, variability is induced in the system. If production rate did not have to be changed to maintain the chest level, the reductions in standard deviations would also be much greater than shown in Table 2.
In an effort to address the first concern, blowline samples were obtained from R2 refiner running in control and manual. Six samples were taken during each mode of operation over an 8 hour period and the samples were sent to an independent lab for quality testing. A summary of these test results is shown in Table 3. Clearly, running the SmartPlates controls in automatic leads to improved pulp quality stability.
Figures 3 and 4 show a snapshot of the operation of the consistency and production controllers, respectively. Each figure shows data for the same 14 hour time period. The reason for showing this "snapshot" is to illustrate the controllers' function – it would not be practical to show a year's worth of data in one figure.
Figure 3 shows that the outlet consistency is kept very constant while the CD dilution water flowrate is manipulated to maintain the consistency. At one point, the operators reduced the flat zone and CD zone gap set-points. Immediately, the consistency increased dramatically. Without a consistency controller to react to this change, the consistency would have remained at the new higher level, which would have caused variability in the refining action and the resulting pulp quality. With the consistency control in automatic, after stabilizing at the new operating conditions the consistency controller made changes to bring the consistency back to the set-point. In other words, the controller is doing what it is supposed to be doing.
Figure 4 shows that the refining zone temperature is kept constant by manipulating the metering screw throughput. The "real mass production rate" is being maintained – as indicated by the constant temperature – by changing the volumetric production rate (metering screw) to compensate for changes in the bulk density of the feed to the refiner. When the operators reduced flat zone and CD zone gaps, the refining zone temperature increased. After stabilizing at a new temperature, the production controller went back to controlling at this new set-point. The production controller is also functioning as it should.
The SmartPlates system at the reference mill was used as a tool to achieve the mill's goals of energy savings and improving pulp quality stability. The system was used for two purposes to achieve these objectives: (1) optimized refiner plate patterns and (2) improved refiner control.
To summarize the key results of running SmartPlates in the rejects refiners:
1. Johansson, O., "Controlling High Consistency Refining Conditions Through Refining Zone Temperature Optimization", 6th PIRA International Refining Conference, Cincinatti, Ohio, March 2001
2. Mosbye, K., Fuglem, G., Kure, K.A., Johansson, O., "Use of Refining Zone Temperature Measurements for Refiner Control", 2001 International Mechanical Pulping Conference, Helsinki, Finland, June 2001
3. Johansson, O., Hogan, D., et. al., "Improved Process Optimization Through Adjustable Refiner Plates", 2001 International Mechanical Pulping Conference, Helsinki, Finland, June 2001