Decision letter for "A survey and comparative evaluation of actor-critic methods in process control"

Process Control Algorithms for Microprocessors

The staking quality of Self-lubricating Spherical Plain Bearings (SSPBs) directly affects the safety of aircraft and the service life of bearings. Reliable loading process control methods and precise process parameter indexes will come into the creation of efficacious staking quality. Therefore, this paper aims to analyze the mechanical state of the roller staking process and give a load control method and corresponding parameter indexes for the high-quality roller staking process. First, based on the analysis of quality inspection requirements, five states of the deformation degree of the flanging lip of the V groove during the roller staking process were proposed, and their relationship with the requirements was studied. Then, the mechanical states corresponding to the five deformation states of the flanging lip deformation were obtained by numerical simulation, and the feeding displacement was determined. Meanwhile, a Multi-Stage Composite Loading (MSCL) process control method was first proposed to control the material damage of the flanging lip, i.e., the rotate speed of the roller tool was constant during the roller staking process, and the displacement–time control was adopted first; when the staking load reaches a staking value, the force–time control was used to make the staking quality meet the requirements. Finally, the staking quality of the MSCL method was verified though the test. The research shows that the feeding displacement needs to be added to the requirements, and the recommended value is 0.5–0.6 times of the V groove depth. A good surface quality and non-material-damage of the flanging lip is more likely to be obtained by the MSCL process control method. The research reveals the formation mechanism of process deformation, and gives more precise process control indexes. At the same time, it provides a theoretical reference for more reliable technical standards.

Surgeons have been slow to incorporate industrial reliability techniques. Process control methods were applied to surgeon waiting time between cases, and to length of stay (LOS) after colon surgery. Waiting times between surgeries were evaluated by auditing the operating room records of a single hospital over a 1-month period. The medical records of 628 patients undergoing colon surgery over a 5-year period were reviewed. The average surgeon wait time between cases was 53 min, and the busiest surgeon spent 291/2 hr in 1 month waiting between surgeries. Process control charting demonstrated poor overall control of the room turnover process. Average LOS after colon resection also demonstrated very poor control. Mean LOS was 10 days. Weibull's conditional analysis revealed a conditional LOS of 9.83 days. Serious process management problems were identified in both analyses. These process issues are both expensive and adversely affect the quality of service offered by the institution. Process control mechanisms were suggested or implemented to improve these surgical processes. Industrial reliability and quality management tools can easily and effectively identify process control problems that occur on surgical services.

The paper describes the application of statistical process control (SPC) methods to supervise the control of a noisy process. SPC charts form a supervisory unit that monitors process outputs and selectively applies terms of a conventional PID controller. The advantages of this method lie in the ability of the control loop to distinguish between process noise and genuine error signals. This stops the controller reacting unnecessarily reducing actuator work. Results are presented for the control of a steam/water heat exchanger rig and show the effectiveness of this supervisory unit to produce quieter control without creating a lag in the control loop. >

In the last years, demands on high quality products increases dramatically, e.g. in textile industry. In order to produce high functional textiles a numerous of chemical agents have to be applied to the fabrics, in several wet chemical treatments, in the production process. One auxiliary agent is the size. Sizes are colorless, water soluble substances which improve the mechanically stiffness of threads during weaving process. Usually, sizes have to be wash out of the fabric before further processing since they may affect the following production steps. Up to now, offline process control methods e.g. extraction of the size are the state of the art of process control in textile industry. It is obvious, that the time consuming and punctual analytical methods are no more seasonable for an optimal process control. Thus, alternative process control methods were on demand. This paper presents a study of the potential of NIR hyperspectral imaging for in-line analysis in textile technology. Application weights and spatial distribution of sizes on polyester fabric are investigated by NIR hyperspectral imaging. In a preliminary study a calibration to the application weight of the size was prepared and a PLS model was established. This PLS model was applied for the quantitative monitoring of the colorless size across the fabrics. Additionally, contaminants on the textile were visualized by NIR chemical imaging. Thus, NIR hyperspectral imaging is presented as a fast, precise and powerful analytical method which also fulfills the requirements of textile industry.

Rapid advancement over the past decades in nanomanufacturing has led to the realization of a broad range of nanostructures such as nanoparticles, nanotubes, and nanowires. The unique mechanical, chemical, and electrical properties of these nanostructures have made them increasingly desired as key components in industrial and commercial applications. As the geometric dimension of nano-manufactured products is on the sub-micron to nanometer scale, different mechanisms and effects are involved in the nanomanufacturing process as compared to those for macro-scale manufacturing. Although direct measurement methods using atomic force microscopy and electron beam microscopy can determine the dimensions of the nano structure with high accuracy, these methods are not suited for online process control and quality assurance. In comparison, indirect measurement methods analyze in-process parameters as the basis for inferring the dimensional variations in the nano products, thereby enabling online feedback for process control and quality assurance. This paper provides a comprehensive review of relevant indirect measurement methods, starting with their respective working principles, and subsequently discussing their characteristics and applications in terms of two different approaches: data-based and physics-based methods. Relevant mathematical and physics models for each of the methods are summarized, together with the associated effect of key process parameters on the quality of the final product. Based on the comprehensive literature conducted, it was found that: (1) indirect measurement, especially the data-based method, plays a critical role when it comes to online process control and quality assurance in nanomanufacturing, because of the short processing time compared to the direct method, and (2) physics-based method is providing a way to optimize the process set up for desired geometrical dimensions.

For the quality management system of herbal medicines, intermediate and finished products it exists the " short board" effect of methodologies. Based on the concept of process control, new strategies and new methods of the production process quality control had been established with the consideration of the actual production of traditional Chinese medicine an the characteristics of Chinese medicine. Taking Banlangen granule as a practice example, which was effective and widespread application, character identification, determination of index components, chemical fingerprint and biometrics technology were sequentially used respectively to assess the quality of Banlangen herbal medicines, intermediate (water extraction and alcohol precipitation) and finished product. With the transfer rate of chemical information and biological potency as indicators, the effectiveness and transmission of the above different assessments and control methods had been researched. And ultimately, the process quality control methods of Banlangen granule, which were based on chemical composition analysis-biometric analysis, had been set up. It can not only validly solute the current status that there were many manufacturers varying quality of Banlangen granule, but also ensure and enhance its clinical efficacy. Furthermore it provided a foundation for the construction of the quality control of traditional Chinese medicine production process.

The reader, on first reading the chapter title, may wonder why process control methods are being discussed within the environmental monitoring section of the book. Process manufacturing and environmental concerns cannot be easily separated and, in fact, the two are interdependent. More often than not the theory behind the analysers is similar but the sample matrix is different. In the design of the process, full environmental impact must be assessed as this determines the route to manufacture. Generally monitoring techniques internal and external to the process will provide both process and environmental data that are invaluable to efficient and safe manufacture.

Development of an improved on-line chromatographic monitor with new methods for environmental and process control

The negative impact of losses from evaporation results not only in reduction of fuel resources and, consequently, cost of products that are lost, but also in negative ecological effect on the environment. Development of modern calculation techniques, prediction and reduction of evaporation burst of light hydrocarbons in the air based on promising methods for monitoring and process control in operation is considered to be an urgent problem. The object of research supports the tendency of gasoline to evaporation under different operating conditions. The subject of study is calculation methodology of petrol losses caused by evaporation. Hypothesis proposed by the authors is that the loss amount of light hydrocarbons from reservoirs depends on a set of factors affecting the tank and gasoline. The solution to establish the relationship between the current factors occurring during evaporation of light hydrocarbons from gasoline during manufacturing operations is viewed as an actual scientific problem.

4 - Spectroscopic on-line monitoring for process control and safeguarding of radiochemical streams in nuclear fuel reprocessing facilities

Data quality

The paper shows how thermal finite-element analysis (FEA), run parallel to the experimental test phase, was used to optimize the unidirectional solidification process of Co-Cr-Mo tension compression specimens and stems of hip-joint endoprostheses by predicting optimal thermal conditions for different thermal boundary conditions and methods of process control. On the basis of the results obtained by FEA simulation of the Bridgman process, FEA is used to simulate and optimize the power-down method, a promising technique for the production of unidirectionally solidified Co-Cr-Mo hip-joint endoprostheses. Next, these results were used to simulate both unidirectional solidification methods for casting a real hip-joint endoprosthesis stem. FEA shows that it is possible to cast unidirectionally solidified stems by either processing method and that, in both cases, no microstructural defects are to be expected. Tension compression fatigue life tests of Co-Cr-Mo samples showed a fatigue life prolongation by a factor of ten for unidirectionally solidified specimens compared to conventionally cast specimens.

Thick corrosion oxide (Fe3O4) layers containing radioactive metals are generated in a carbon steel heat transport system (HTS) of pressurized heavy water reactors during operation. It is necessary to decontaminate the oxide layers to reduce radiation exposure to workers before maintenance and decommissioning. Since the oxide layer thickness in the HTS is about 75 μm, a considerable amounts of secondary wastes can be generated when the existing decontamination processes are applied. The aim of this study is to design a chemical decontamination process using chelate-free inorganic acid that can be effectively applied to remove the oxide layers in the HTS and to minimize the secondary waste generation. Therefore, Fe3O4 dissolution behaviors were evaluated in a Hydrazine Based Reductive metal Ion Decontamination (HyBRID) process using the chelate-free inorganic acid. From this result, the decontamination process of oxide layers in the HTS using recycling of decontamination process solution and the process control method using pH monitoring were constructed. In particular, process control based on pH monitoring can be easily used because decontamination performance can be controlled according to the goal; (1) effective removing the oxide layer at decommissioning, (2) avoiding corrosion of base metal and removing the oxide layer at maintenance.

Questions about the validity of measurements made in an E/sup 3/ laboratory using software to control the instrumentation are a continuing concern for EMC engineers. In previous work, it has been shown that the use of statistical process control (SPC) methods to gage and track the variability of system measurements can provide confidence in the repeatability of the system hardware setup, and in the data it produces. Once the hardware system stability has been demonstrated, it is possible to also utilize SPC historical information to assist in confirming the validity of the data gathered using the system's control software.