Kitchen waste recycling control parameter optimization system

Abstract

The invention belongs to the technical field of kitchen waste treatment, and discloses a kitchen waste recycling control parameter optimization system which comprises the following steps: collecting kitchen waste treatment parameters in the same time sequence, and preprocessing the kitchen waste treatment parameters to obtain standard kitchen waste parameters; the standard kitchen waste parameters are predicted based on the constructed microorganism parameter prediction model, microorganism parameters are obtained, microorganism activity evaluation is conducted on the microorganism parameters, and the growth rate is obtained; performing gas production evaluation on the standard kitchen waste parameters to obtain a gas production rate; constructing a resource balance function based on the growth rate and the gas production rate, and performing parameter optimization on the resource balance function to obtain an optimal parameter adjustment combination; and the treatment efficiency and the recycling output rate of kitchen waste treatment are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of kitchen waste treatment, and more specifically, to a kitchen waste resource utilization control parameter optimization system. Background Art

[0002] In the process of food waste treatment, the activity and gas production efficiency of microorganisms are the key factors affecting the resource utilization effect. Traditional optimization methods often cannot effectively balance the activity and gas production efficiency of microorganisms. For example, in the treatment of high-salt and high-fat food waste, high salt and high fat content not only inhibit the growth of microorganisms, but also may lead to a decrease in gas production efficiency, thereby affecting the resource utilization efficiency. Traditional optimization methods are often based on static parameter settings, and fail to effectively consider environmental changes and dynamic changes in the growth process of microorganisms, and cannot effectively balance the activity and gas production efficiency of microorganisms. There is a certain balance between microbial activity and gas production efficiency. Efficient waste resource utilization requires that microorganisms can reproduce rapidly in a suitable environment and effectively decompose organic matter in the waste to produce gas (such as methane). However, the activity and gas production efficiency of microorganisms are usually affected by multiple factors, including temperature, pH, moisture content and organic matter content. Under certain working conditions, these factors may interact with each other. Too high temperature or pH value may inhibit the activity of microorganisms, while too low temperature may lead to a decrease in gas production efficiency, making it difficult for the activity and gas production efficiency of microorganisms to reach the optimal state at the same time.

[0003] In view of this, the present invention proposes a food waste resource utilization control parameter optimization system to solve the above problems. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a system for optimizing control parameters for resource utilization of kitchen waste, comprising:

[0005] Data collection module: collects the restaurant kitchen waste treatment parameters of the same time series, pre-processes the restaurant kitchen waste treatment parameters, and obtains standard restaurant kitchen waste parameters;

[0006] Microbial evaluation module: including a prediction unit and an evaluation unit. The prediction unit predicts the standard food waste parameters based on the constructed microbial parameter prediction model to obtain the microbial parameters. The evaluation unit evaluates the microbial activity of the microbial parameters to obtain the growth rate.

[0007] Gas production evaluation module: evaluate the gas production of standard kitchen waste parameters and obtain the gas production rate;

[0008] Parameter optimization module: construct a resource balance function based on the growth rate and gas production rate, tune the parameters of the resource balance function, and obtain the best parameter adjustment combination.

[0009] Further, the kitchen waste treatment parameters include: ambient temperature, pH value, moisture content, and organic content.

[0010] Further, the method for preprocessing the kitchen waste treatment parameters includes:

[0011] Presetting a filter and the cut-off frequency of the filter, inputting the kitchen waste treatment parameters into the set filter, and the filter separates the frequency components of the ambient temperature, pH value, and moisture content in the kitchen waste treatment parameters, filters out the high-frequency noise above the cut-off frequency, and obtains the filtered parameters; using an anomaly detection algorithm to identify anomalies in the filtered parameters, marking the filtered parameters identified as anomalies as anomalies, removing the filtered parameters marked as anomalies in the filtered parameters, and the remaining filtered parameters constitute the effective kitchen waste treatment parameters; constructing a non-linear standard mapping function, and the formula of the non-linear standard mapping function is: where Y represents the standard mapping value, X represents the input parameter, and a1, a2, and a3 represent the mapping parameters; using each data in the effective kitchen waste treatment parameters as the input parameter of the non-linear standard mapping function for calculation to obtain the standard mapping value, and all the standard mapping values constitute the standard kitchen waste parameters.

[0012] Further, the method for constructing the microbial parameter prediction model includes:

[0013] Collecting M groups of historical kitchen waste treatment parameters and historical microbial parameters, the historical kitchen waste treatment parameters are of the same data type as the data in the kitchen waste treatment parameters, and the historical microbial parameters include: lag phase and decay coefficient; using the historical kitchen waste treatment parameters as the input parameters of the non-linear standard mapping function for calculation to obtain the historical standard kitchen waste parameters;

[0014] Based on the historical standard kitchen waste parameters and historical microbial parameters, using the linear regression model as the initial model of the microbial parameter prediction model, using the historical standard kitchen waste parameters and historical microbial parameters as training data, using the training data as the training sample set, training the linear regression model with the training sample set, using the historical standard kitchen waste parameters as the input data of the microbial parameter prediction model, and using the predicted microbial parameters as the output data of the microbial parameter prediction model; taking the minimization of the error between the actual historical microbial parameters and the microbial parameters predicted by the microbial parameter prediction model as the training objective, using the mean square error function as the loss function of the microbial parameter prediction model, and stopping training to obtain the microbial parameter prediction model when the loss function converges.

[0015] Further, the formula for evaluating the microbial activity of the microbial parameters is: where μ represents the growth rate, Cre maxrepresents the maximum growth rate, t represents the time scale corresponding to the standard kitchen waste treatment parameters, γ represents the lag period in the microbial parameters, e represents the Euler number, and β represents the decay coefficient in the microbial parameters.

[0016] Further, the method for gas production evaluation of the standard kitchen waste parameters includes:

[0017] Collect group A evaluation training data, where the evaluation training data includes: historical kitchen waste treatment parameters and the corresponding gas production rates; use the historical kitchen waste treatment parameters as the input parameters of the non-linear standard mapping function for calculation to obtain historical standard treatment parameters, and use each group of historical standard treatment parameters and the corresponding gas production rates as data samples. All data samples form a sample training set;

[0018] Preset a bias perturbation factor and a Lagrange multiplier perturbation factor, and initialize the optimal particle as empty; preset B particles, each particle represents a parameter combination, and each parameter combination includes: data bias, a feature data set, and a set of Lagrange multipliers corresponding to the feature data set, and preset an initial gas production evaluation model; for each particle, use the sample training set to perform particle evaluation on each particle to obtain the particle fitness; based on the particle fitness of each particle, select the particle with the smallest particle fitness as the particle to be selected, compare the particle fitness of the particle to be selected with that of the optimal particle, and when the particle fitness of the particle to be selected is less than that of the optimal particle, use the particle to be selected as the new optimal particle; based on the bias perturbation factor and the Lagrange multiplier perturbation factor, perform state update on each particle, add a bias perturbation factor to the data bias in the particle, and add a Lagrange multiplier perturbation factor to each Lagrange multiplier in the set of Lagrange multipliers in the particle, and repeat until the optimal particle no longer changes, and output the optimal particle at this time as the best parameter combination;

[0019] Put the best parameter combination into the initial gas production evaluation model to obtain the optimal gas production evaluation model, and use the standard kitchen waste parameters as the input data of the optimal gas production evaluation model for calculation to obtain the gas production rate.

[0020] Further, the formula of the initial gas production evaluation model is:

[0021] where Q represents the predicted gas production rate, N represents the size of the feature data set, α i represents the Lagrange multiplier corresponding to the i-th data in the feature data set, α i * represents the dual multiplier corresponding to the i-th data in the feature data set, Da i represents the i-th data in the feature data set, In represents the input data, b represents the data bias, c represents the offset constant, and d represents the dimension parameter.

[0022] Furthermore, the formula for evaluating each particle is as follows:

[0023] where Fit j represents the particle fitness of the j-th particle, Q true,e represents the gas production rate of the e-th data in the sample training set, Q pre,e represents the predicted gas production rate of the e-th data in the sample training set, Size represents the total amount of data in the sample training set, C represents the penalty factor, and Pun(Q true,e , Q pre,e ) represents the penalty function.

[0024] Furthermore, the formula for the resource balance function is: Sous = w1×μ + w2×Q; where Sous represents the resource evaluation, w1 represents the microbial activity weight, and w2 represents the gas production rate weight; the microbial activity weight and the gas production rate weight satisfy the resource weight constraint: w1 + w2 = 1.

[0025] Furthermore, the methods for parameter tuning of the resource balance function include:

[0026] Step 1: Preset the fitness threshold and the optimal chromosome, and initialize the optimal chromosome as empty;

[0027] Step 2: Preset K groups of random adjustment parameter combinations. The adjustment parameter combinations include the adjustment parameters for each parameter in the standard food waste parameters. Each group of adjustment parameter combinations is represented as a chromosome, and the genes in the chromosome correspond to the adjustment parameters in the adjustment parameter combination;

[0028] Step 3: For each chromosome, calculate the resource evaluation value corresponding to each chromosome according to the resource balance function, use the resource evaluation value as the fitness of each chromosome, and select the parental chromosome through the tournament selection algorithm according to the fitness value;

[0029] Step 4: According to the parental chromosome, preset the U-th gene as the exchange node, exchange the genes of the parental chromosome at the exchange node to generate the offspring chromosome;

[0030] Step 5: Preset the mutation perturbation factor, and through the random mutation algorithm, randomly select a gene of the offspring chromosome to increase or decrease a mutation perturbation factor to obtain the mutant chromosome;

[0031] Step 6: Calculate the fitness of the mutated chromosome according to the resource balance function, and select the mutated chromosome with the maximum fitness as the preselected chromosome. When the optimal chromosome is empty or the fitness value of the preselected chromosome is greater than the fitness of the optimal chromosome, the preselected chromosome is used as the new optimal chromosome; when the maximum iteration number is not reached and the fitness value of the optimal chromosome is less than the preset fitness threshold, return to Step 3 to continue the iteration; when the maximum iteration number is reached or the fitness value of the optimal chromosome is greater than the preset fitness threshold, stop the iteration and output the optimal chromosome as the best parameter adjustment combination.

[0032] Technical effects and advantages of the optimized system for control parameters of kitchen waste resource utilization in the present invention:

[0033] By preprocessing the parameters of kitchen waste, the present invention improves the quality and accuracy of data, ensures the comparability of treatment parameters in different time periods and environments, and eliminates data deviation; by constructing a prediction model for microbial parameters, it can accurately predict key microbial parameters and improve the accuracy of microbial activity evaluation; by evaluating the microbial activity of microbial parameters, it takes into account the dynamic influence of environmental parameters on microbial activity, thus being more in line with the actual scenario; by constructing an initial gas production evaluation model and selecting the optimal parameter combination through the particle swarm optimization algorithm, it ensures the high accuracy and applicability of gas production rate prediction, enabling it to adapt to the treatment requirements of different batches of kitchen waste; by constructing a resource balance function, comprehensively considering the weights of microbial activity and gas production rate, and introducing a genetic algorithm to optimize the parameters of the resource balance function, it ensures the global search ability and fast convergence characteristics of the optimization process; the system is applicable to the treatment scenarios of kitchen waste with various environmental temperatures, pH values, moisture contents and organic contents, significantly improves the treatment efficiency, has a positive impact on the economy of energy recovery, reduces the environmental pollution of untreated waste and waste and unnecessary resource consumption in the treatment process, and promotes green development. Description of the Drawings

[0034] Figure 1 It is a schematic diagram of an optimized system for control parameters of kitchen waste resource utilization in the present invention;

[0035] Figure 2 It is a schematic diagram of an optimized method for control parameters of kitchen waste resource utilization in the present invention. Detailed Embodiments

[0036] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

[0037] Example 1

[0038] Please refer to Figure 1 As shown, a control parameter optimization system for the resource utilization of kitchen waste in this embodiment includes:

[0039] Data acquisition module: Collect the kitchen waste treatment parameters in the same time series, preprocess the kitchen waste treatment parameters, and obtain the standard kitchen waste parameters;

[0040] Microbial evaluation module: including a prediction unit and an evaluation unit. The prediction unit predicts the standard kitchen waste parameters based on the constructed microbial parameter prediction model to obtain microbial parameters. The evaluation unit conducts a microbial activity evaluation on the microbial parameters to obtain the growth rate;

[0041] Gas production evaluation module: Conduct a gas production evaluation on the standard kitchen waste parameters to obtain the gas production rate;

[0042] Parameter optimization module: Construct a resource utilization balance function based on the growth rate and the gas production rate, and optimize the parameters of the resource utilization balance function to obtain the best parameter adjustment combination;

[0043] The kitchen waste treatment parameters include: environmental temperature, pH value, moisture content, and organic content; the environmental temperature is obtained through a temperature sensor; the pH value represents the concentration of acidic and alkaline substances in the kitchen waste and is obtained through a pH sensor; the moisture content represents the proportion of water in the kitchen waste and is a key parameter affecting microbial activity and the fermentation process, and is obtained through a humidity sensor; the organic content represents the proportion of biodegradable organic matter in the kitchen waste and is obtained through a near-infrared spectrometer;

[0044] Each module is connected by wired and / or wireless means to achieve data transmission between modules.

[0045] During the kitchen waste treatment process, the data collected by the sensors are often affected by factors such as environmental noise and equipment vibration. These high-frequency noises may affect the subsequent analysis accuracy. The filter can effectively remove the high-frequency noises and retain the key low-frequency signals; the original data in the kitchen waste treatment usually has non-linear characteristics. Directly using the original data will lead to large errors in subsequent processing. Through non-linear mapping standardization, the inconsistencies and scale differences in the data can be eliminated, the data quality can be improved, and the stability and comparability of the data under different working conditions can be ensured, making the subsequent calculation of microbial activity and gas production efficiency more accurate. Specifically:

[0046] A preset filter and the cut-off frequency of the filter are set. The kitchen waste treatment parameters are input into the set filter. The filter separates the frequency components of the environmental temperature, pH value, and moisture content in the kitchen waste treatment parameters, filters out the high-frequency noise above the cut-off frequency, and obtains the filtered parameters. Common filters include low-pass filters and mean filters. An anomaly detection algorithm is used to identify anomalies in the filtered parameters. The filtered parameters identified as anomalies are marked as anomalies, and the filtered parameters marked as anomalies in the filtered parameters are removed. The remaining filtered parameters constitute the effective kitchen waste treatment parameters. Commonly used anomaly detection algorithms include the standard score method and the median absolute deviation method. A non-linear standard mapping function is constructed. The formula of the non-linear standard mapping function is: Where Y represents the standard mapping value, X represents the input parameter, and a1, a2, and a3 represent the mapping parameters, which are used to define the behavior of the non-linear standard mapping function. The non-linear standard mapping function has strong non-linear characteristics, can effectively compress a wide range of numerical values, limit the range of the output results, and make it within a suitable interval. Each data in the effective kitchen waste treatment parameters is used as the input parameter of the non-linear standard mapping function for calculation to obtain the standard mapping value. All the standard mapping values constitute the standard kitchen waste parameters.

[0047] During the fermentation process of microorganisms, especially under special working conditions (such as high-fat and high-salt environments), the growth rate changes are relatively complex. Traditional growth models may not be able to accurately describe this complex relationship. By evaluating the microbial activity of microbial parameters, considering the lag phase, maximum growth rate, and decay coefficient of microorganisms, the estimation of microbial activity is made more accurate, thus providing a more accurate basis for optimizing treatment parameters. Specifically:

[0048] Collect M groups of historical kitchen waste treatment parameters and historical microbial parameters. The historical kitchen waste treatment parameters are of the same data type as the data in the kitchen waste treatment parameters. The historical microbial parameters include the lag phase and the decay coefficient. The lag phase represents the time interval from when the microorganisms start to grow. Microorganisms under different kitchen waste treatment parameters have different lag phases. The decay coefficient represents the decay rate of microbial growth. Microorganisms under different kitchen waste treatment parameters have different decay coefficients. The historical kitchen waste treatment parameters are used as the input parameters of the non-linear standard mapping function for calculation to obtain the historical standard kitchen waste parameters;

[0049] Based on historical standard kitchen waste parameters and historical microbial parameters, a linear regression model is used as the initial model of the microbial parameter prediction model. The historical standard kitchen waste parameters and historical microbial parameters are used as training data, and the training data is used as the training sample set. The linear regression model is trained using the training sample set. The historical standard kitchen waste parameters are used as the input data of the microbial parameter prediction model, and the predicted microbial parameters are used as the output data of the microbial parameter prediction model. The training objective is to minimize the error between the actual historical microbial parameters and the microbial parameters predicted by the microbial parameter prediction model. The mean square error function is used as the loss function of the microbial parameter prediction model. When the loss function converges, the training stops to obtain the microbial parameter prediction model.

[0050] Based on the constructed microbial parameter prediction model, the standard kitchen waste parameters are used as the input of the microbial parameter prediction model for prediction to obtain the microbial parameters.

[0051] Microbial activity assessment is performed on the microbial parameters. The formula for microbial activity assessment of the microbial parameters is: where μ represents the growth rate, Cre max represents the maximum growth rate. The maximum growth rate is the maximum growth rate per unit time of the microorganism under the most favorable environmental conditions, which is the maximum potential for microbial growth and is obtained through corresponding experiments by those skilled in the art under the optimal temperature and optimal nutritional conditions. t represents the time scale corresponding to the standard kitchen waste treatment parameters, γ represents the lag phase in the microbial parameters, e represents the Euler number (the base of the natural logarithm), and β represents the decay coefficient in the microbial parameters.

[0052] During the fermentation process of kitchen waste, the gas production rate is an important indicator for evaluating the resource utilization efficiency. Due to the non-linearity and complexity of the fermentation process itself, traditional linear regression models are difficult to accurately capture the non-linear relationships therein, thus affecting the prediction effect of the gas production rate and further affecting the optimization of the entire treatment process. By performing gas production assessment on the standard kitchen waste parameters, an optimal gas production assessment model is constructed to enable the system to efficiently and accurately predict the gas production rate during the fermentation process of kitchen waste. Specifically:

[0053] Collect group A of evaluation training data. The evaluation training data includes: historical kitchen waste treatment parameters and the corresponding gas production rates. The historical kitchen waste treatment parameters are of the same data type as the data within the kitchen waste treatment parameters. The gas production rate is the gas generation rate at different times during the garbage treatment process under fixed kitchen waste treatment parameters and is obtained by those skilled in the art based on records during the actual garbage treatment process. The historical kitchen waste treatment parameters are used as the input parameters of the non-linear standard mapping function for calculation to obtain the historical standard treatment parameters. Each set of historical standard treatment parameters and the corresponding gas production rates are used as data samples, and all data samples constitute the sample training set.

[0054] Preset a bias perturbation factor and a Lagrange multiplier perturbation factor, and initialize the optimal particle as empty; preset B particles, each particle represents a parameter combination, and each parameter combination includes: data bias, a feature dataset, and a corresponding Lagrange multiplier set for the feature dataset; there is a one-to-one mapping relationship between the feature dataset and the Lagrange multiplier set, that is, each data in the feature dataset has a unique corresponding Lagrange multiplier in the Lagrange multiplier set; preset an initial gas production evaluation model, and the formula of the initial gas production evaluation model is: where Q represents the predicted gas production rate, N represents the size of the feature dataset, α i represents the Lagrange multiplier corresponding to the i-th data in the feature dataset, α i * represents the dual multiplier corresponding to the i-th data in the feature dataset, Da i represents the i-th data in the feature dataset, In represents the input data, b represents the data bias, which is used to control the prediction error, c represents the offset constant, which is used to control the offset, and d represents the dimension parameter, which is used to control the mapping dimension;

[0055] For each particle, use the sample training set to evaluate each particle, and the formula for evaluating each particle is:

[0056] where Fit j represents the particle fitness of the j-th particle, Q true,e represents the gas production rate of the e-th data in the sample training set, Q pre,e represents the predicted gas production rate of the e-th data in the sample training set, Size represents the total amount of data in the sample training set, C represents the penalty factor, Pun(Q true,e ,Q pre,e ) represents the penalty function. When the deviation between the predicted gas production rate and the actual gas production rate of the e-th data in the sample training set is greater than the preset deviation threshold, the value of the penalty function is 1, otherwise it is zero; the particle fitness represents the deviation between the predicted gas production rate and the actual gas production rate. The smaller the particle fitness, the better the performance of the parameter combination corresponding to the particle for the initial gas production evaluation model; based on the particle fitness of each particle, select the particle with the smallest particle fitness as the particle to be selected, compare the particle fitness of the particle to be selected with that of the optimal particle, and when the particle fitness of the particle to be selected is less than that of the optimal particle, use the particle to be selected as the new optimal particle; update the state of each particle based on the bias perturbation factor and the Lagrange multiplier perturbation factor, add a bias perturbation factor to the data bias in the particle, and add a Lagrange multiplier perturbation factor to each Lagrange multiplier in the Lagrange multiplier set of the particle. Repeat until the optimal particle no longer changes, and output the optimal particle at this time as the best parameter combination;

[0057] The initial gas production evaluation model is placed with the optimal parameter combination to obtain the optimal gas production evaluation model. Using the standard kitchen waste parameters as the input data of the optimal gas production evaluation model for calculation, the gas production rate is obtained.

[0058] During the kitchen waste treatment process, there is a coupling relationship between multiple control parameters. Single-objective optimization often fails to take into account all factors, resulting in low resource utilization efficiency. By optimizing the parameters of the resource utilization objective function, comprehensive optimization of multiple control objectives (microbial activity and gas production efficiency) can be achieved, effectively balancing various indicators, and thus improving the resource utilization efficiency. Specifically:

[0059] A resource utilization balance function is constructed based on the gas production rate and the growth rate. The formula of the resource utilization balance function is: Sous = w1×μ + w2×Q; where Sous represents the resource utilization evaluation, w1 represents the microbial activity weight, w2 represents the gas production rate weight. When higher microbial activity is required, a larger microbial activity weight can be set. If more attention is paid to the resource utilization efficiency (i.e., gas production efficiency), a larger gas production rate weight can be set; the microbial activity weight and the gas production rate weight satisfy the resource weight constraint:

[0060] w1 + w2 = 1; Preset the target gas production rate weight and the target microbial activity weight, substitute the target gas production rate weight and the target microbial activity weight into the resource utilization balance function, and optimize the parameters of the resource utilization balance function:

[0061] Step 1: Preset the fitness threshold and the optimal chromosome, and initialize the optimal chromosome as empty;

[0062] Step 2: Preset K groups of random adjustment parameter combinations. The adjustment parameter combination includes the adjustment parameters (i.e., adjustment amounts) for each parameter in the standard kitchen waste parameters. Each group of adjustment parameter combinations is represented as a chromosome, and the genes in the chromosome correspond to the adjustment parameters in the adjustment parameter combination;

[0063] Step 3: For each chromosome, calculate the resource utilization evaluation value corresponding to each chromosome according to the resource utilization balance function, use the resource utilization evaluation value as the fitness of each chromosome, and select the parental chromosomes through the tournament selection algorithm according to the fitness value;

[0064] Step 4: According to the parental chromosomes, preset the Uth gene as the exchange node, exchange the genes of the parental chromosomes at the exchange node to generate offspring chromosomes;

[0065] Step 5: Preset the mutation perturbation factor, and through the random mutation algorithm, randomly select a gene of the offspring chromosome to increase or decrease a mutation perturbation factor to obtain the mutated chromosome;

[0066] Step 6: Calculate the fitness of the mutant chromosomes according to the resource balance function, and select the mutant chromosome with the highest fitness as the preselected chromosome. When the optimal chromosome is empty or the fitness value of the preselected chromosome is greater than that of the optimal chromosome, the preselected chromosome is used as the new optimal chromosome; when the maximum iteration number is not reached and the fitness value of the optimal chromosome is less than the preset fitness threshold, return to Step 3 to continue the iteration; when the maximum iteration number is reached or the fitness value of the optimal chromosome is greater than the preset fitness threshold, stop the iteration and output the optimal chromosome as the best parameter adjustment combination.

[0067] In this embodiment, by preprocessing the parameters of food waste, the quality and accuracy of the data are improved, ensuring the comparability of the processing parameters in different time periods and environments, and eliminating data deviation; by constructing a microbial parameter prediction model, key microbial parameters can be accurately predicted, improving the accuracy of microbial activity evaluation; by evaluating the microbial activity of microbial parameters, the dynamic influence of environmental parameters on microbial activity is considered, thus being more in line with the actual scenario; by constructing an initial gas production evaluation model and selecting the optimal parameter combination through the particle swarm optimization algorithm, the high accuracy and applicability of gas production rate prediction are ensured, enabling it to adapt to the processing requirements of different batches of food waste; by constructing a resource balance function, comprehensively considering the weights of microbial activity and gas production rate, and introducing a genetic algorithm to optimize the parameters of the resource balance function, the global search ability and fast convergence characteristics of the optimization process are ensured; the system is applicable to food waste treatment scenarios with various environmental temperatures, pH values, moisture contents, and organic contents, significantly improving the treatment efficiency, having a positive impact on the economy of energy recovery, reducing the environmental pollution of untreated waste and waste and unnecessary resource consumption during the treatment process, and promoting green development.

[0068] Embodiment 2;

[0069] Please refer to Figure 2 As shown, for the parts not described in detail in this embodiment, refer to the description content of Embodiment 1. A method for optimizing the resource control parameters of food waste is provided, including:

[0070] S1. Collect the food waste treatment parameters in the same time series, preprocess the food waste treatment parameters, and obtain the standard food waste parameters;

[0071] S2. Based on the constructed microbial parameter prediction model, predict the standard food waste parameters to obtain microbial parameters, and evaluate the microbial activity of the microbial parameters to obtain the growth rate;

[0072] S3. Conduct gas production evaluation on the standard food waste parameters to obtain the gas production rate;

[0073] S4. Construct a resource balance function based on the growth rate and gas production rate, optimize the parameters of the resource balance function, and obtain the optimal parameter adjustment combination.

[0074] Example 3;

[0075] This embodiment publicly provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the operation mode of the above-provided method for optimizing the control parameters of food waste resource utilization.

[0076] Since the electronic device introduced in this embodiment is the electronic device used to implement the method for optimizing the control parameters of food waste resource utilization in the embodiments of the present application, based on the method for optimizing the control parameters of food waste resource utilization introduced in the embodiments of the present application, those skilled in the art can understand the specific implementation manners and various forms of change of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device implements the method in the embodiments of the present application will not be described in detail here. As long as those skilled in the art implement the electronic device used in the method for optimizing the control parameters of food waste resource utilization in the embodiments of the present application, it belongs to the scope of protection of the present application.

[0077] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0078] The above are only the preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for ordinary technical users in the technical field, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of the present invention.

Claims

1. A system for optimizing control parameters of resource utilization of kitchen waste, characterized in that, Including: Data acquisition module: It acquires the kitchen waste treatment parameters in the same time series, preprocesses the kitchen waste treatment parameters, and obtains the standard kitchen waste parameters; Microbial evaluation module: It includes a prediction unit and an evaluation unit. The prediction unit predicts the standard kitchen waste parameters based on the constructed microbial parameter prediction model to obtain microbial parameters. The evaluation unit conducts a microbial activity evaluation on the microbial parameters to obtain the growth rate; Gas production evaluation module: It conducts a gas production evaluation on the standard kitchen waste parameters to obtain the gas production rate; Parameter optimization module: It constructs a resource balance function based on the growth rate and the gas production rate, tunes the parameters of the resource balance function, and obtains the optimal parameter adjustment combination.

2. The optimized system for control parameters of kitchen waste resource utilization according to claim 1, characterized in that, The kitchen waste treatment parameters include: environmental temperature, pH value, moisture content, and organic content.

3. The optimized system for control parameters of kitchen waste resource utilization according to claim 2, characterized in that, The method for preprocessing the kitchen waste treatment parameters includes: Preset a filter and the cut-off frequency of the filter, input the kitchen waste treatment parameters into the set filter. The filter separates the frequency components of the environmental temperature, pH value, and moisture content in the kitchen waste treatment parameters, filters out the high-frequency noise above the cut-off frequency, and obtains the filtered parameters. Use an anomaly detection algorithm to identify anomalies in the filtered parameters, mark the filtered parameters identified as anomalies, remove the marked filtered parameters from the filtered parameters, and the remaining filtered parameters constitute the effective kitchen waste treatment parameters. Construct a non-linear standard mapping function, and the formula of the non-linear standard mapping function is: where Y represents the standard mapping value, X represents the input parameter, and a1, a2, and a3 represent the mapping parameters; calculate using each data in the effective kitchen waste treatment parameters as the input parameter of the non-linear standard mapping function to obtain the standard mapping value, and all the standard mapping values constitute the standard kitchen waste parameters.

4. The optimized system for control parameters of kitchen waste resource utilization according to claim 3, characterized in that, The method for constructing the microbial parameter prediction model includes: Collect M groups of historical kitchen waste treatment parameters and historical microbial parameters. The historical kitchen waste treatment parameters are of the same data type as the data in the kitchen waste treatment parameters. The historical microbial parameters include: lag phase and decay coefficient. Calculate using the historical kitchen waste treatment parameters as the input parameters of the non-linear standard mapping function to obtain the historical standard kitchen waste parameters; Based on the historical standard kitchen waste parameters and historical microbial parameters, use the linear regression model as the initial model of the microbial parameter prediction model. Use the historical standard kitchen waste parameters and historical microbial parameters as training data, use the training data as the training sample set, train the linear regression model using the training sample set, use the historical standard kitchen waste parameters as the input data of the microbial parameter prediction model, and use the predicted microbial parameters as the output data of the microbial parameter prediction model. Use minimizing the error between the actual historical microbial parameters and the microbial parameters predicted by the microbial parameter prediction model as the training objective, use the mean square error function as the loss function of the microbial parameter prediction model. When the loss function reaches convergence, stop training to obtain the microbial parameter prediction model.

5. The optimized system for control parameters of kitchen waste resource utilization according to claim 4, wherein The formula for evaluating the microbial activity of microbial parameters is as follows: where μ represents the growth rate, Cre max represents the maximum growth rate, t represents the time scale corresponding to the standard kitchen waste treatment parameters, γ represents the lag phase in the microbial parameters, e represents the Euler number, and β represents the decay coefficient in the microbial parameters.

6. The optimized system for control parameters of kitchen waste resource utilization according to claim 5, characterized in that, The method for conducting a gas production evaluation on the standard kitchen waste parameters includes: Collect A groups of evaluation training data. The evaluation training data includes: historical kitchen waste treatment parameters and the corresponding gas production rates. Calculate using the historical kitchen waste treatment parameters as the input parameters of the non-linear standard mapping function to obtain the historical standard treatment parameters. Use each group of historical standard treatment parameters and the corresponding gas production rates as data samples, and all data samples form a sample training set; Preset a bias perturbation factor and a Lagrange multiplier perturbation factor, and initialize the optimal particle as empty; preset B particles, each particle represents a parameter combination, and each parameter combination includes: data bias, a feature data set, and a corresponding Lagrange multiplier set of the feature data set, and preset an initial gas production evaluation model; for each particle, use the sample training set to evaluate each particle to obtain the particle fitness; based on the particle fitness of each particle, select the particle with the minimum particle fitness as the particle to be selected, compare the particle fitness of the particle to be selected with that of the optimal particle, and when the particle fitness of the particle to be selected is less than that of the optimal particle, use the particle to be selected as the new optimal particle; update the state of each particle based on the bias perturbation factor and the Lagrange multiplier perturbation factor, add a bias perturbation factor to the data bias in the particle, and add a Lagrange multiplier perturbation factor to each Lagrange multiplier in the Lagrange multiplier set of the particle, and repeat until the optimal particle no longer changes, and output the optimal particle at this time as the best parameter combination; Place the best parameter combination into the initial gas production evaluation model to obtain the optimal gas production evaluation model, and use the standard kitchen waste parameters as the input data of the optimal gas production evaluation model for calculation to obtain the gas production rate.

7. The optimized system for control parameters of resource utilization of kitchen waste according to claim 6, wherein The formula of the initial gas production evaluation model is: Among them, Q represents the predicted gas production rate, N represents the size of the characteristic data set, and α i represents the Lagrange multiplier corresponding to the i-th data in the characteristic data set, and α i * represents the dual multiplier corresponding to the i-th data in the characteristic data set, Da i represents the i-th data in the characteristic data set, In represents the input data, b represents the data bias, c represents the offset constant, and d represents the dimension parameter.

8. The optimized system for control parameters of kitchen waste resource utilization according to claim 7, characterized in that, The formula for evaluating each particle is: Among them, Fit j represents the particle fitness of the j-th particle, Q true,e represents the gas production rate of the e-th data in the sample training set, Q pre,e represents the predicted gas production rate of the e-th data in the sample training set, Size represents the total amount of data in the sample training set, C represents the penalty factor, Pun(Q true,e ,Q pre,e ) represents the penalty function.

9. The optimized system for control parameters of kitchen waste resource utilization according to claim 8, characterized in that, The formula of the resource balance function is: Sous = w1×μ + w2×Q; where, Sous represents the resource evaluation, w1 represents the microbial activity weight, and w2 represents the gas production rate weight; the microbial activity weight and the gas production rate weight satisfy the resource weight constraint: w1 + w2 = 1.

10. The optimized system for control parameters of kitchen waste resource utilization according to claim 8, wherein The methods for parameter tuning of the resource balance function include: Step 1, preset a fitness threshold and an optimal chromosome, and initialize the optimal chromosome as empty; Step 2, preset K groups of random adjustment parameter combinations, and the adjustment parameter combination includes the adjustment parameters for each parameter in the standard kitchen waste parameters. Each group of adjustment parameter combinations is represented as a chromosome, and the genes in the chromosome correspond to the adjustment parameters in the adjustment parameter combination; Step 3, for each chromosome, calculate the resource evaluation value corresponding to each chromosome according to the resource balance function, use the resource evaluation value as the fitness of each chromosome, and select the parental chromosome through the tournament selection algorithm according to the value of the fitness; Step 4, according to the parental chromosome, preset the U-th gene as the exchange node, exchange the genes of the parental chromosome at the exchange node to generate the offspring chromosome; Step 5, preset a mutation perturbation factor, and through the random mutation algorithm, randomly select a gene of the offspring chromosome to add or subtract a mutation perturbation factor to obtain the mutant chromosome; Step 6: Calculate the fitness of the mutated chromosome according to the resource balance function, and select the mutated chromosome with the maximum fitness as the preselected chromosome. When the optimal chromosome is empty or the fitness value of the preselected chromosome is greater than the fitness of the optimal chromosome, use the preselected chromosome as the new optimal chromosome; when the maximum iteration number has not been reached and the fitness value of the optimal chromosome is less than the preset fitness threshold, return to Step 3 to continue the iteration; when the maximum iteration number is reached or the fitness value of the optimal chromosome is greater than the preset fitness threshold, stop the iteration and output the optimal chromosome as the best parameter adjustment combination.

Images