A PET plastic manufacturing automated processing method and system

Abstract

The invention discloses an automated processing method and system for PET plastic manufacturing, and specifically relates to the field of automated control technology, comprising the following steps: in an automated processing project, an automated cooling control module for controlling cooling equipment sends different types of control instructions to the cooling equipment according to cooling requirements, evaluates the response to the cooling requirements according to the response time after the control instructions are sent and received, and determines whether there is a hidden danger of invalid instructions; if there is a hidden danger of invalid instructions, the operation data and instruction response data of the cooling equipment are acquired and feature extraction is performed, and then fuzzy logic reasoning is performed according to the feature extraction result to determine whether the control instructions of the cooling equipment need to be resent, and the optimization type when resending; the invention can determine the response efficiency of the equipment, timely detect and eliminate the hidden dangers of hysteresis, delay and invalid instructions, ensure that the equipment can execute instructions quickly and accurately, and ensure the fluency of the production process.

Description

Technical Field

[0001] The present invention relates to the field of automatic control technology, and more specifically, to an automatic processing method and system for PET plastic manufacturing. Background Art

[0002] With the rapid development of industrial automation, the demand for control of precision equipment in the production process is increasing, especially in manufacturing processes involving high temperature sensitivity. How to ensure the operating efficiency of cooling equipment, reduce energy consumption and ensure production stability has become an important issue. The manufacturing process of PET plastics is particularly dependent on precise temperature control to ensure the stability of product quality. However, traditional cooling systems are usually unable to effectively regulate according to the dynamic needs of the production process, which may lead to insufficient or excessive cooling, thereby affecting production efficiency, increasing energy consumption, and even causing substandard product quality. In addition, due to the complex operating environment of the equipment, the inconsistency between command transmission and equipment response time often leads to inefficient equipment operation and increased maintenance and operating costs. Therefore, how to optimize the response effect of cooling equipment to control instructions has become a technical problem facing the current industrial automation cooling system. Summary of the invention

[0003] To achieve the above object, the present invention provides the following technical solutions:

[0004] A PET plastic manufacturing automated processing method comprises the following steps:

[0005] In automated processing projects, the automated cooling control module that controls the cooling equipment sends different types of control instructions to the cooling equipment according to the cooling demand, evaluates the response to the cooling demand based on the response time after the control instruction is sent and received, and determines whether there is a hidden danger of invalid instructions;

[0006] If there is a hidden danger of invalid instructions, the operating data and instruction response data of the cooling equipment are acquired and feature extraction is performed. Then, fuzzy logic reasoning is performed based on the feature extraction results to determine whether the control instructions of the cooling equipment need to be resent and the type of optimization when resending.

[0007] In a preferred embodiment, each type of control instruction corresponds to a change combination of data corresponding to a control parameter of the cooling device.

[0008] In a preferred embodiment, evaluating the response to the cooling demand based on the response time after the control instruction is sent and received refers to:

[0009] In a fixed time window, obtain the timestamps when all types of control instructions are sent and the timestamps when the cooling device receives the control instruction and responds to it. According to the sending order of the control instructions, obtain the time series X, and calculate the average value and standard deviation of the time series X respectively.

[0010] In a preferred embodiment, determining whether there is a hidden danger of invalid instructions refers to:

[0011] The average value and standard deviation of the time series X are obtained, and then the average value of the time series X is compared with a preset first reference threshold, and the standard deviation of the time series X is compared with a preset second reference threshold. If the average value of the time series X is less than or equal to the preset first reference threshold and the standard deviation of the time series X is less than or equal to the preset second reference threshold, a normal signal is generated; if the average value of the time series X is less than or equal to the preset first reference threshold and the standard deviation of the time series X is less than or equal to the preset second reference threshold is not satisfied, an abnormal signal is generated, and when the abnormal signal is generated, it indicates that there is a hidden danger of invalid instructions.

[0012] In a preferred embodiment, performing feature extraction refers to:

[0013] Extract power consumption matching features from the operation data and generate a power consumption matching index;

[0014] Response matching features are extracted from instruction response data and a response matching index is generated.

[0015] In a preferred implementation, the logic for generating the power consumption matching index is:

[0016] In the feature extraction time window, the actual power consumption data P1(t) of each time period is obtained from the cooling device in real time, and the corresponding time point t is recorded to obtain the expected power consumption P2(t) of the device at each time point. The error ΔP(t) between the actual power consumption and the expected power consumption at each time point t is calculated:

[0017] ΔP(t)=P1(t)-P2(t); then the power consumption deviation in the entire feature extraction time window, i.e., the time series of the error ΔP(t), is obtained, and the average value μ1 and standard deviation σ1 of the power consumption deviation are calculated respectively;

[0018] Use the time warping algorithm to analyze the match between actual power consumption and expected power consumption:

[0019] Align the time series of actual power consumption data P1(t) and the time series of expected power consumption P2(t), and calculate the matching degree corresponding to the optimal time alignment path:

[0020] DTW(P1, P2) = min∑ (i，j)∈Wd[P1(i), P2(j)]; W represents the time alignment path, d[P1(i), P2(j)] represents the distance between the time series of the actual power consumption data P1(t) at time i and the time series of the expected power consumption data P2(t) at time j, and DTW(P1, P2) represents the matching degree of the two time series;

[0021] Then substitute it into the following formula to calculate the power consumption matching index:

[0022] PMI=1-α*|μ1|-β*σ1-γ*DTW(P1, P2); PMI represents the power consumption matching index, and α, β, and γ are all preset proportional coefficients.

[0023] In a preferred embodiment, d[P1(i), P2(j)] takes the Euclidean distance.

[0024] In a preferred embodiment, the generation logic of the response matching index is:

[0025] In the time window of feature extraction, obtain the time series X and the standard time series X' corresponding to the time series X, and calculate the offset time series of the two time series:

[0026] ΔT(t)=x1(t)-x2(t); ΔT(t) represents the offset time value corresponding to time t in the offset time series, x1(t) represents the data corresponding to time t in the time series X, and x2(t) represents the data corresponding to time t in the standard time series X'. The mean value μ2 and standard deviation σ2 of the offset time series are calculated respectively;

[0027] Use cross-correlation analysis to evaluate the similarity between time series X and a standard time series X':

[0028] μ x1 represents the mean value of the time series X, μ x2 represents the average value of the standard time series X', τ represents the time lag, R(τ) is the cross-correlation function, and N represents the length of the time series X;

[0029] The maximum value Rmax of the cross-correlation function is used as a measure of the similarity between the time series X and the standard time series X';

[0030] Then substitute it into the following formula to calculate the response matching index:

[0031] RMI=Rmax-f1*|μ2|-f2*σ2; RMI represents the response matching index, and f1 and f2 are both preset influence coefficients.

[0032] In a preferred embodiment, fuzzy logic reasoning is performed to determine whether the control instruction of the cooling device needs to be resent, and the optimization type when resending refers to:

[0033] The power consumption matching index, response matching index, and whether the demand has changed are taken as input variables, and whether the control instructions of the cooling equipment need to be resent and the optimization type when resending are taken as output variables. The input variables are fuzzified and the values ​​of the input variables are converted into fuzzy sets. The output variables are fuzzified and the output variables are converted into fuzzy sets. Fuzzy rules are formulated to describe the resending requirements under different data type combinations and the optimization type requirements when resending. The fuzzified input variables are inferred through fuzzy rules to obtain whether the control instructions of the cooling equipment need to be resent and the optimization type when resending.

[0034] In a preferred embodiment, a PET plastic manufacturing automated processing system comprises:

[0035] Automatic cooling control module, used to send different types of control instructions to cooling equipment according to cooling requirements in automated processing projects;

[0036] The control and analysis module evaluates the response to the cooling demand based on the response time after the control command is sent and received, and determines whether there is a hidden danger of invalid commands;

[0037] The feature extraction module acquires the operating data and the command response data of the cooling equipment and performs feature extraction to obtain feature extraction results if there is a hidden danger of invalid instructions.

[0038] The optimization reasoning module performs fuzzy logic reasoning based on the feature extraction results to determine whether the control instructions of the cooling equipment need to be resent and the optimization type when resending.

[0039] Technical effects and advantages of the present invention:

[0040] The present invention realizes real-time response to cooling demand through the automatic cooling control module, and makes intelligent judgments based on the operating status, power consumption matching index and response matching index of the equipment, ensuring that the cooling equipment can be dynamically adjusted according to the actual situation. This greatly improves the intelligence level of the system, reduces human intervention, and ensures the automation and efficiency of the cooling process.

[0041] The present invention can detect and optimize the power output of cooling equipment in real time through accurate power consumption matching and response time evaluation, and automatically adjust the power, flow rate and temperature according to actual needs, thus avoiding unnecessary energy waste. The use of power consumption matching index can ensure the maximum operating efficiency of equipment, achieve continuous optimization of energy efficiency, and reduce production costs.

[0042] By evaluating the sending and response time of instructions, the present invention can determine the response efficiency of the equipment, timely detect and eliminate the hidden dangers of lag, delay and invalid instructions. Fuzzy logic reasoning further improves the system's response speed and decision-making accuracy under complex working conditions, ensuring that the equipment can execute instructions quickly and accurately, and ensuring the smoothness of the production process.

[0043] Since the cooling process directly affects the quality of the finished PET plastic product, the present invention ensures that the cooling process will not be overcooled or undercooled through intelligent regulation. The equipment can dynamically adjust the cooling parameters according to changes in production conditions, thereby stabilizing the cooling effect and improving the consistency and yield rate of the product.

[0044] By extracting the features of power consumption and response data, the present invention can detect potential equipment failures in advance, adjust the equipment operation strategy in time, and avoid losses caused by improper equipment operation. This preventive maintenance reduces equipment downtime and failure rate, extends the service life of the equipment, and reduces maintenance and replacement costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0046] Figure 1 The schematic diagram is a schematic diagram of an automated processing method for manufacturing PET plastics in the present invention.

[0047] Figure 2 The schematic diagram of an automated processing system for manufacturing PET plastics in the present invention is shown in FIG. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Reference Figure 1 - Figure 2 The following examples are obtained:

[0050] Embodiment 1: A method for automated processing of PET plastics, comprising the following steps:

[0051] In automated processing projects, the automated cooling control module that controls the cooling equipment sends different types of control instructions to the cooling equipment according to the cooling demand, evaluates the response to the cooling demand based on the response time after the control instruction is sent and received, and determines whether there is a hidden danger of invalid instructions;

[0052] The automatic cooling control module automatically adjusts the operation of the equipment according to the cooling demand, which not only improves the efficiency of the cooling process, but also ensures that the plastic will not have quality problems during the molding process due to insufficient or excessive cooling. It obtains data such as temperature and power consumption in real time, and sends appropriate control instructions (such as adjusting the cooling medium flow, equipment power, etc.) to meet different cooling needs. The automatic cooling control module has a built-in preset automatic control strategy to respond to different needs.

[0053] The cooling demand is not constant, it will fluctuate with changes in production conditions (such as ambient temperature, production speed). The automatic cooling control module sends different types of instructions according to the cooling demand to ensure that the cooling equipment is always in the best operating state, avoiding energy waste or product quality problems caused by overcooling or insufficient cooling. The automatic cooling control module will dynamically send appropriate control instructions according to current needs to adjust parameters such as cooling power, flow rate or wind speed.

[0054] Whether the actual response of the equipment is timely directly affects the cooling effect. Evaluating the sending and receiving time of the command is the key to ensuring the accurate response of the system. It helps to identify whether there are problems with lag, error or invalid instructions. By comparing the sending time of the control instruction and the response time of the equipment, the system can determine whether the instruction is executed effectively. If the response is too slow or there is no response, there may be a device problem or communication failure.

[0055] If the hidden danger of invalid instructions is found, it may cause deviations in the cooling process and affect the quality of the finished product. Detecting and dealing with these hidden dangers in advance can avoid system errors. Feature extraction is to deeply analyze the actual operation of the equipment. By extracting the key features of power consumption and response time of the cooling equipment, we can better understand the operating status of the equipment and provide accurate data support for subsequent judgments.

[0056] If there is a hidden danger of invalid instructions, the operating data and instruction response data of the cooling equipment are acquired and feature extraction is performed. Then, fuzzy logic reasoning is performed based on the feature extraction results to determine whether the control instructions of the cooling equipment need to be resent and the type of optimization when resending.

[0057] Since the state changes of cooling equipment are complex and cannot be accurately quantified, fuzzy logic can help the system make accurate judgments under uncertain conditions. By analyzing the fuzzy set of equipment response data, the system can intelligently decide whether to take action. Fuzzy logic reasoning will analyze the results of feature extraction and determine whether the equipment needs to be adjusted according to the set rules. If the system detects that the equipment is not operating properly or fails to respond as expected, it is necessary to determine whether to resend the instruction. Resending can avoid insufficient cooling or overcooling caused by equipment failure or instruction transmission problems. Through fuzzy logic reasoning, the system can determine whether the current control instruction is invalid, and if invalid, decide to resend the instruction. When resending instructions, the system should not simply repeat the previous instructions, but optimize the type of resending according to the current cooling needs and equipment status. This can ensure that the operation of the cooling equipment is more precise, reduce energy consumption and improve cooling efficiency.

[0058] Each type of control instruction corresponds to a change combination of data of a cooling device control parameter, for example, the flow rate of the coolant increases from 1000 L / min to 1500 L / min.

[0059] Evaluating the response to cooling demand based on the response time after the control command is sent and received means: in a fixed time window, obtaining the timestamps when all types of control commands are sent and the timestamps when the cooling device receives the control command and responds, and obtaining the time series X according to the order in which the control commands are sent, and calculating the average value and standard deviation of the time series X respectively.

[0060] Judging whether there is a hidden danger of invalid instructions refers to: obtaining the average value and standard deviation of the time series X, and then comparing the average value of the time series X with a preset first reference threshold, and comparing the standard deviation of the time series X with a preset second reference threshold. If the average value of the time series X is less than or equal to the preset first reference threshold and the standard deviation of the time series X is less than or equal to the preset second reference threshold, a normal signal is generated. If the average value of the time series X is less than or equal to the preset first reference threshold and the standard deviation of the time series X is less than or equal to the preset second reference threshold, an abnormal signal is generated. When the abnormal signal is generated, it indicates that there is a hidden danger of invalid instructions.

[0061] Collecting the sending and response time of control instructions within a specified time window can provide basic data for subsequent analysis. Selecting a fixed time window can help the system determine the overall response efficiency and stability of the cooling equipment within a certain time range, avoiding misjudgments caused by local data. Record the sending time of each control instruction and the response time of the cooling equipment. Through these timestamp data, the system can determine the response of the equipment to different instructions at different times.

[0062] The time series X represents the time difference between sending and receiving all control commands. By calculating the mean and standard deviation of the time series, it can be determined whether the response of the cooling equipment is stable and efficient.

[0063] Mean value: indicates the average response time of the cooling device to all commands. If the mean value is small, it means that the device responds quickly. Standard deviation: indicates the fluctuation of the response time. If the standard deviation is small, it means that the device response time is stable and the fluctuation is small. The mean value can characterize the overall response speed of the device, while the standard deviation characterizes the consistency and fluctuation of the response. They provide key statistics for subsequent judgments. By comparing the mean value and standard deviation of the time series X with the preset benchmark threshold, the system can determine whether there is a problem of slow or unstable device response. If the set threshold conditions are not met, it means that the cooling device may not effectively respond to certain control commands, and there is a risk of invalid commands. The first benchmark threshold: is used to evaluate whether the response time of the cooling device is too long. If the average response time exceeds this threshold, it means that the device is lagging behind and there is a risk of invalid commands. The second benchmark threshold: is used to evaluate the fluctuation of the response time. If the standard deviation exceeds this threshold, it means that the device response is inconsistent and some commands may not be effectively executed.

[0064] Feature extraction refers to: extracting power consumption matching features from operating data and generating a power consumption matching index; extracting response matching features from instruction response data and generating a response matching index.

[0065] The power matching index is used to measure the difference between the actual power consumption of the cooling equipment and the expected power consumption. If the actual power consumption of the cooling equipment deviates from the expected, it may mean that the equipment is operating in an inefficient manner, resulting in energy waste or substandard cooling effect. By calculating the power matching index, the system can detect these deviations in time and optimize them to ensure that the equipment maintains a balance between energy consumption and cooling requirements.

[0066] Abnormal fluctuations in power consumption may also reflect potential problems with the cooling equipment (such as mechanical failure or control system failure), which requires maintenance or repair. Therefore, the power matching index PMI is an important indicator of the health status of the equipment. By comparing the actual power consumption with the expected power consumption, PMI can reflect whether the equipment is operating efficiently according to system instructions. When the index is low, it means that the equipment may be inefficient or unstable and needs to be adjusted.

[0067] The response matching index (RMI) is used to evaluate the response speed and execution effect of the cooling equipment after receiving the instruction, and to measure the degree of match between the actual response time and the expected time. The response time of the equipment is directly related to the effective execution of the control instruction. If the equipment response is delayed or inaccurate, the cooling equipment may not be able to meet the demand in time, affecting the smooth progress of the production process. Through RMI, the system can accurately evaluate the execution effect of the equipment's instructions. In complex automation systems, cooling needs and instructions change frequently. RMI can promptly identify problems with delayed or inconsistent equipment responses, thereby ensuring the system's real-time control capabilities under dynamic changes. RMI can reflect the speed and accuracy of the equipment's response to control instructions by calculating the degree of match between the actual response and the expected response time series. A high RMI indicates that the equipment responds quickly and accurately to the control instruction, otherwise it may indicate that the equipment is lagging or the response is unstable.

[0068] The power consumption matching index focuses on the efficiency of the power consumption of the equipment during actual operation, ensuring that the cooling process meets the demand without generating excessive energy waste. The response matching index focuses on the equipment's ability to respond to control instructions, ensuring that the equipment can execute instructions in a timely and effective manner, and ensuring that the cooling system meets changing needs in real time.

[0069] The generation logic of the power consumption matching index is:

[0070] In the feature extraction time window, the actual power consumption data P1(t) of each time period is obtained from the cooling device in real time, and the corresponding time point t is recorded to obtain the expected power consumption P2(t) of the device at each time point. The error ΔP(t) between the actual power consumption and the expected power consumption at each time point t is calculated:

[0071] ΔP(t)=P1(t)-P2(t); then the power consumption deviation in the entire feature extraction time window, i.e., the time series of the error ΔP(t), is obtained, and the average value μ1 of the power consumption deviation is calculated to reflect the overall deviation between the actual power consumption and the expected power consumption in the entire cooling cycle, and the standard deviation σ1 is calculated to reflect the magnitude of the power consumption fluctuation. A large fluctuation indicates that the system is unstable.

[0072] Use the time warping algorithm to analyze the match between actual power consumption and expected power consumption:

[0073] Align the time series of actual power consumption data P1(t) and the time series of expected power consumption P2(t), and calculate the matching degree corresponding to the optimal time alignment path:

[0074] DTW(P1, P2) = min∑ (i，j)∈Wd[P1(i), P2(j)]; W represents the time alignment path, d[P1(i, P2(j)] represents the distance between the time series of the actual power consumption data P1(t) at time i and the time series of the expected power consumption data P2(t) at time j, DTW(P1, P2) represents the matching degree of the two time series, the calculation result of the dynamic time warping algorithm, which is used to evaluate the time alignment of the two power consumption series. The smaller the value, the higher the matching degree.

[0075] Then substitute it into the following formula to calculate the power consumption matching index:

[0076] PMI = 1-α*|μ1|-β*σ1-γ*DTW(P1, P2); PMI represents the power consumption matching index, α, β, and γ are preset proportional coefficients, which are used to adjust the influence of average error, variance, and time series alignment on PMI. DTW measures the matching degree between two time series. The smaller the value, the closer the actual power consumption is to the expected power consumption in terms of timing. 1-α*|μ1|-β*σ1 measures the size and fluctuation of power consumption deviation. The smaller the power consumption deviation, the more stable the device operation and the better the power consumption matching, so the value of this part is closer to 1. DTW can handle nonlinear changes in power consumption over time and is an effective measure of the time dynamic matching of actual power consumption and expected power consumption. The closer the PMI index is to 1, the better the actual power consumption matches the expected power consumption, and the closer the device operation is to the expected state. On the contrary, the lower the PMI, the larger the power consumption deviation or the unstable device response, and the poorer the matching degree.

[0077] d[P1(i), P2(j)] takes the Euclidean distance. For time series, the Euclidean distance calculates the difference in power consumption data at two time points. Specifically: The Euclidean distance can intuitively reflect the gap between two values. It is suitable for the comparison of continuous values ​​such as power consumption data. In the comparison of power consumption time series, the Euclidean distance can effectively measure the difference in power consumption at different time points, reflecting the deviation between the actual operation of the device and the expected operation. In addition, the Euclidean distance in the DTW algorithm helps to solve the timing problem in time series matching, especially when the power consumption changes nonlinearly over time. By finding the best matching path, the Euclidean distance can accurately evaluate the difference between the two sequences.

[0078] The generation logic of the response matching index is:

[0079] In the time window of feature extraction, obtain the time series X and the standard time series X' corresponding to the time series X, and calculate the offset time series of the two time series:

[0080] ΔT(t)=x1(t)-x2(t); ΔT(t) represents the offset time value corresponding to time t in the offset time series, x1(t) represents the data corresponding to time t in the time series X, and x2(t) represents the data corresponding to time t in the standard time series X'. The mean value μ2 and standard deviation σ2 of the offset time series are calculated respectively;

[0081] Use cross-correlation analysis to evaluate the similarity between time series X and a standard time series X':

[0082] μ x1 represents the mean value of the time series X, μ x2 represents the average value of the standard time series X', τ represents the time lag, which is used to align the time difference between two time series, R(τ) is the cross-correlation function, which is used to measure the similarity between the time series X and the standard time series X' under the time lag τ, and N represents the length of the time series X;

[0083] The maximum value Rmax of the cross-correlation function is used as a measure of the similarity between the time series X and the standard time series X'; the maximum value Rmax indicates that the maximum cross-correlation value represents the best alignment point between the time series X and the standard time series X'. If the Rmax value is high, it means that the two sequences are very well matched in time, and the system's response to the command is almost consistent with expectations. μ2 and σ2 are used to measure the overall deviation and fluctuation of the time offset, respectively. A higher μ2 indicates that the average offset of the sequence in time is larger, while a higher σ2 indicates that the fluctuation of the response time is larger. By incorporating these factors into the response matching index, the stability and consistency of the system response can be more comprehensively evaluated.

[0084] Substitute into the following formula to calculate the response matching index:

[0085] RMI=Rmax-f1*|μ2|-f2*σ2; RMI stands for response matching index, where f1 and f2 are preset influence coefficients used to adjust the weight of average offset and fluctuation. A high RMI value indicates that the actual response time of the device is very close to the expected response time, the matching effect is good, and the device response is stable. A low RMI value indicates that there may be problems with the response of the device, such as the response time deviates from expectations or there are large fluctuations.

[0086] Fuzzy logic reasoning is performed to determine whether the control instructions of the cooling equipment need to be resent, and the optimization type when resending refers to:

[0087] The power consumption matching index, response matching index, and whether the demand has changed are taken as input variables, and whether the control instructions of the cooling equipment need to be resent and the optimization type when resending are taken as output variables. The input variables are fuzzified and the values ​​of the input variables are converted into fuzzy sets. The output variables are fuzzified and the output variables are converted into fuzzy sets. Fuzzy rules are formulated to describe the resending requirements under different data type combinations and the optimization type requirements when resending. The fuzzified input variables are inferred through fuzzy rules to obtain whether the control instructions of the cooling equipment need to be resent and the optimization type when resending.

[0088] Input variables: Power Match Index (PMI): Indicates the degree of match between the actual power consumption of the device and the expected power consumption. Response Match Index (RMI): Indicates the response time and accuracy of the device to instructions. Whether the demand changes: Indicates the dynamic changes in cooling demand. If the demand changes, the control instructions may need to be adjusted.

[0089] Fuzzy processing: convert the values ​​of the three input variables into fuzzy sets. For example: PMI: can be divided into "good", "medium" and "low". RMI: can be divided into "fast response", "medium" and "slow response". Demand change: can be divided into "no change" and "change".

[0090] Output variables: Whether to resend the command: Controls whether the command needs to be resent. Optimization type when resending: Whether to optimize the communication channel quality, the equipment operation quality, both, or neither when resending.

[0091] Optimize the communication channel: If the device is detected to be slow to respond to instructions or has communication delays, optimizing the communication channel can improve the efficiency of data transmission between the device and the control system. Optimize device operation: Improve the operating efficiency and responsiveness of the device by adjusting the power, flow or working mode of the device. Optimize both: Optimize the communication channel and device operation at the same time to ensure that the device can respond and execute quickly and accurately after receiving instructions. Do not optimize either: If the system determines that the current status and conditions meet the requirements, there is no need to optimize, and the current operating status can be maintained.

[0092] Fuzzification: whether to resend the instruction: can be fuzzified into "resend" and "no need to resend". Optimization type: can be fuzzified into "optimize communication channel", "optimize device operation", "optimize both", "do not optimize either".

[0093] By formulating fuzzy rules, different combinations of input variables are associated with output variables. The following are examples of possible fuzzy rules:

[0094] Rule 1: If PMI is poorly matched and RMI is slow to respond, and there are changes in requirements, it is necessary to resend instructions and optimize communication channels and equipment operations.

[0095] Rule 2: If PMI is a high match and RMI is a fast response, and there is no change in demand, there is no need to resend instructions and no optimization is required.

[0096] Rule 3: If the PMI is a moderate match and the RMI is a moderate response, and there is a change in demand, resend the instruction and optimize the equipment operation.

[0097] Rule 4: If PMI is a low match, but RMI is a fast response, and there is no change in demand, resend the instruction and optimize the communication channel.

[0098] These rules describe how to decide whether the instruction needs to be resent and the specific optimization plan based on different input conditions. After the input variables are fuzzified, reasoning is performed through pre-established fuzzy rules. Based on the input of power consumption matching index, response matching index and demand change, the system uses fuzzy rules to determine whether the instruction needs to be resent and determines the type of optimization. After the output result is obtained through fuzzy reasoning, it is necessary to perform defuzzification to convert the fuzzy set into a clear decision value. Through fuzzy logic reasoning, the system can intelligently determine whether the control instructions of the cooling equipment need to be resent, and which aspects (communication channel or equipment operation quality) should be optimized when resending. This method combines multiple factors of power consumption matching index, response matching index and demand change to ensure that the cooling system maintains efficient and precise regulation in dynamic changes.

[0099] Embodiment 2: A PET plastic manufacturing automated processing system, comprising:

[0100] Automatic cooling control module, used to send different types of control instructions to cooling equipment according to cooling requirements in automated processing projects;

[0101] The control and analysis module evaluates the response to the cooling demand based on the response time after the control command is sent and received, and determines whether there is a hidden danger of invalid commands;

[0102] The feature extraction module acquires the operating data and the command response data of the cooling equipment and performs feature extraction to obtain feature extraction results if there is a hidden danger of invalid instructions.

[0103] The optimization reasoning module performs fuzzy logic reasoning based on the feature extraction results to determine whether the control instructions of the cooling equipment need to be resent and the optimization type when resending.

[0104] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0105] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0106] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0107] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0108] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A PET plastic manufacturing automated processing method, characterized in that: The following steps are involved: In automated processing projects, the automated cooling control module that controls the cooling equipment sends different types of control instructions to the cooling equipment according to the cooling demand, evaluates the response to the cooling demand based on the response time after the control instruction is sent and received, and determines whether there is a hidden danger of invalid instructions; If there is a hidden danger of invalid instructions, the operation data and instruction response data of the cooling equipment are acquired and feature extraction is performed, and then fuzzy logic reasoning is performed based on the feature extraction results to determine whether the control instructions of the cooling equipment need to be resent and the optimization type when resending; Feature extraction refers to: Extract power consumption matching features from the operation data and generate a power consumption matching index; Extract response matching features from instruction response data and generate a response matching index; The generation logic of the power consumption matching index is: In the feature extraction time window, the actual power consumption data P1(t) of each time period is obtained from the cooling device in real time, and the corresponding time point t is recorded to obtain the expected power consumption P2(t) of the device at each time point. The error ΔP(t) between the actual power consumption and the expected power consumption at each time point t is calculated: ΔP(t)=P1(t)-P2(t); then the power consumption deviation in the entire feature extraction time window, i.e., the time series of the error ΔP(t), is obtained, and the average value μ1 and standard deviation σ1 of the power consumption deviation are calculated respectively; Use the time warping algorithm to analyze the match between actual power consumption and expected power consumption: Align the time series of actual power consumption data P1(t) and the time series of expected power consumption P2(t), and calculate the matching degree corresponding to the optimal time alignment path: DTW(P1, P2)=minΣ (i，j)∈W d[P1(i), P2(j)]; W represents the time alignment path, d[P1(i), P2(j)] represents the distance between the time series of the actual power consumption data P1(t) at time i and the time series of the expected power consumption data P2(t) at time j, and DTW(P1, P2) represents the matching degree of the two time series; Then substitute it into the following formula to calculate the power consumption matching index: PMI = 1-α*|μ1|-β*σ1-γ*DTW(P1, P2); PMI represents the power consumption matching index, and α, β, and γ are all preset proportional coefficients; The generation logic of the response matching index is: In the time window of feature extraction, obtain the time series X and the standard time series X' corresponding to the time series X, and calculate the offset time series of the two time series: ΔT(t)=x1(t)-x2(t); ΔT(t) represents the offset time value corresponding to time t in the offset time series, x1(t) represents the data corresponding to time t in the time series X, and x2(t) represents the data corresponding to time t in the standard time series X'. The mean value μ2 and standard deviation σ2 of the offset time series are calculated respectively; Use cross-correlation analysis to evaluate the similarity between time series X and a standard time series X': μ x1 represents the mean value of the time series X, μ x2 represents the average value of the standard time series X', τ represents the time lag, R(τ) is the cross-correlation function, and N represents the length of the time series X; The maximum value Rmax of the cross-correlation function is used as a measure of the similarity between the time series X and the standard time series X'; Then substitute it into the following formula to calculate the response matching index: RMI=Rmax-f1*|μ2|-f2*σ2; RMI represents the response matching index, and f1 and f2 are both preset influence coefficients.

2. The automated processing method for manufacturing PET plastic according to claim 1, characterized in that: Each type of control instruction corresponds to a change combination of data of a cooling device control parameter.

3. A PET plastic manufacturing automated processing method according to claim 2, characterized in that: Evaluating the response to cooling demand based on the response time after control commands are sent and received refers to: In a fixed time window, obtain the timestamps when all types of control instructions are sent and the timestamps when the cooling device receives the control instruction and responds to it. According to the sending order of the control instructions, obtain the time series X, and calculate the average value and standard deviation of the time series X respectively.

4. The automated processing method for manufacturing PET plastic according to claim 3, characterized in that: The hidden dangers of judging whether there are invalid instructions are: The average value and standard deviation of the time series X are obtained, and then the average value of the time series X is compared with a preset first reference threshold, and the standard deviation of the time series X is compared with a preset second reference threshold. If the average value of the time series X is less than or equal to the preset first reference threshold and the standard deviation of the time series X is less than or equal to the preset second reference threshold, a normal signal is generated; if the average value of the time series X is less than or equal to the preset first reference threshold and the standard deviation of the time series X is less than or equal to the preset second reference threshold is not satisfied, an abnormal signal is generated, and when the abnormal signal is generated, it indicates that there is a hidden danger of invalid instructions.

5. The automated processing method for manufacturing PET plastic according to claim 4, characterized in that: d[P1(i), P2(j)] takes the Euclidean distance.

6. The automated processing method for manufacturing PET plastic according to claim 5, characterized in that: Fuzzy logic reasoning is performed to determine whether the control instructions of the cooling equipment need to be resent, and the optimization type when resending refers to: The power consumption matching index, response matching index, and whether the demand has changed are taken as input variables, and whether the control instructions of the cooling equipment need to be resent and the optimization type when resending are taken as output variables. The input variables are fuzzified and the values ​​of the input variables are converted into fuzzy sets. The output variables are fuzzified and the output variables are converted into fuzzy sets. Fuzzy rules are formulated to describe the resending requirements under different data type combinations and the optimization type requirements when resending. The fuzzified input variables are inferred through fuzzy rules to obtain whether the control instructions of the cooling equipment need to be resent and the optimization type when resending.

7. A PET plastic manufacturing automated processing system, implemented based on the PET plastic manufacturing automated processing method according to any one of claims 1 to 6, characterized in that: include: Automatic cooling control module, used to send different types of control instructions to cooling equipment according to cooling requirements in automated processing projects; The control and analysis module evaluates the response to the cooling demand based on the response time after the control command is sent and received, and determines whether there is a hidden danger of invalid commands; The feature extraction module acquires the operating data and the command response data of the cooling equipment and performs feature extraction to obtain feature extraction results if there is a hidden danger of invalid instructions. The optimization reasoning module performs fuzzy logic reasoning based on the feature extraction results to determine whether the control instructions of the cooling equipment need to be resent and the optimization type when resending.

Images