Building concrete waste treatment system

Abstract

The invention discloses a building concrete waste treatment system, which belongs to the technical field of building waste treatment, and is characterized in that a data acquisition module acquires waste physical attributes and inputs the waste physical attributes into a model generation and optimization module; the model generation and optimization module establishes a waste physical characteristic model, dynamically updates the waste physical characteristic model and outputs a dynamic waste physical characteristic model to the strategy generation module; the strategy generation module outputs equipment operation parameters; the strategy mapping module maps the equipment operation parameters into physical execution signals through an equipment instruction mapping model to obtain execution results, and the execution results are input into the strategy feedback optimization module; and the strategy feedback optimization module identifies the execution deviation in the execution result and feeds the execution deviation back to the strategy generation module to optimize the equipment operation parameters. According to the scheme, an intelligent algorithm and physical modeling are combined, the waste treatment process is optimized through real-time data collection and feedback, equipment parameters are adjusted based on different characteristics of the waste, and the problems of excessive treatment and resource waste are avoided.

Description

Technical Field

[0001] The invention belongs to the technical field of construction waste treatment, and in particular relates to a construction concrete waste treatment system. Background Art

[0002] The construction industry has always produced a large amount of construction waste, especially concrete waste, and its recycling and treatment is a major challenge in the current management of construction waste. With the continuous expansion of construction projects and the demolition of old buildings, the amount of construction waste has also continued to increase. Especially in the process of urbanization, the accumulation of construction waste has brought huge environmental pressure. As the most common type of construction waste, concrete waste is difficult to handle, mainly reflected in the classification, recycling and resource utilization of waste. At present, most of the treatment of concrete waste still relies on traditional manual classification, physical treatment (such as crushing, screening, etc.) and mechanical sorting technologies. However, these traditional technologies have many shortcomings in practical applications, resulting in low treatment efficiency, low recovery rate and prominent environmental problems.

[0003] Among the existing waste treatment methods, although manual sorting can sort the waste to a certain extent, the efficiency of manual sorting is extremely low and errors are prone to occur due to the complex composition of concrete waste and the difficulty in identifying it quickly and accurately. Although mechanized treatment can improve efficiency, it still faces problems such as inaccurate sorting, excessive energy consumption, and low degree of automation. In particular, mixed materials and pollutants in waste usually require human intervention for manual screening, which not only wastes a lot of human resources, but also results in a large amount of recyclable resources not being effectively recycled. Although some current high-tech treatment methods (such as the use of laser scanning, spectral analysis and other technologies) have made certain technological progress, they have not been widely used due to the high cost of equipment, complex operation and high level of maintenance required. Furthermore, the common problems of existing technologies are that they lack intelligence and adaptability and cannot be adjusted and optimized in real time according to the different types of waste and changing conditions.

[0004] Furthermore, existing waste treatment technologies face environmental challenges. Traditional waste recycling processes generate significant levels of exhaust gas, noise, and dust pollution. However, waste still contains a large amount of recyclable components (such as sand and gravel in concrete). Failure to efficiently and accurately recycle these components will inevitably lead to significant waste of resources, further exacerbating the environmental burden. Therefore, improving waste recycling efficiency, reducing energy consumption, and achieving intelligent waste sorting and processing have become core issues that require urgent solutions.

[0005] To this end, we propose a construction concrete waste treatment system to solve the above problems. Summary of the Invention

[0006] The purpose of the present invention is to solve the problem that the existing technology cannot recycle efficiently and accurately, and to propose a construction concrete waste treatment system.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A construction concrete waste treatment system, comprising:

[0009] A data acquisition module, which is used to collect physical properties of waste materials to obtain initial data and input the initial data into the model generation and optimization module;

[0010] A model generation and optimization module, which is used to establish a waste material physical property model based on the initial data collected by the data acquisition module, and to update and modify the waste material physical property model through a closed-loop feedback mechanism to output a dynamic waste material physical property model;

[0011] a strategy generation module, the strategy generation module being configured to use the dynamic waste physical property model output by the model generation and optimization module as an environmental state input and output equipment operating parameters of the waste treatment equipment based on the optimization strategy;

[0012] A strategy mapping module maps device operating parameters into physical execution signals through a device instruction mapping model, sends them to corresponding devices, obtains execution results, and inputs the execution results into a strategy feedback optimization module;

[0013] The strategy feedback optimization module identifies the execution deviation of the device operating parameters in the execution result, and feeds back the execution deviation to the strategy generation module to optimize the device operating parameters to obtain the optimized device operating parameters.

[0014] Preferably, the initial data collected by the data collection module includes: humidity, particle size, temperature and density of the waste.

[0015] Preferably, a corresponding weight is assigned to each physical property in the initial data in the model generation and optimization module to establish a waste physical property model. The waste physical property model is dynamically updated according to the data collected by the data acquisition module at different times. The weight is mainly adjusted based on the importance of each physical property to the processing effect, and a regularization term is set to smooth the changes in the waste physical property model.

[0016] Preferably, the physical property model P of the waste in the model generation and optimization module f is represented as follows:

[0017]

[0018] Where: wi is each physical property X i The weight, w i satisfy X i is the i-th physical property value of the waste.

[0019] Preferably, the device operating parameters of the policy generation module are generated by the following policy:

[0020] S t =π θ (P f (t))+β·Ω(P f (t))

[0021] Among them, S t is the set of equipment control parameters of the material handling system; θ (P f (t)) is based on the physical property model P f Optimization strategy for the physical properties of (t); Ω(P f (t)) is based on the physical property model P f The physical feasibility correction term of the physical characteristics of the waste (t) performs physical boundary correction on the output equipment parameters according to the current physical state of the waste to ensure that the strategy results conform to the physical laws; β is the adjustment coefficient of the physical correction term, which is used to balance the flexibility and physical rationality of strategy generation.

[0022] Preferably, the optimization strategy in the strategy generation module is based on the adoption strategy S t After the recovery efficiency index and the use of S t The energy consumption indicators of the post-processing equipment are optimized by balancing the relationship between the physical properties of the waste, recycling efficiency and energy consumption.

[0023] Preferably, the policy feedback optimization module uses a historical sliding window regularization mechanism to balance short-term response and long-term convergence in policy updates, specifically:

[0024]

[0025] Where, ΔS t : Policy deviation, used to guide π θ Adjustment; Recovery rate targets set by the system; Actual treatment recovery rate; E t / Q t : Unit energy consumption; D var 、M var : The particle size fluctuation rate and humidity fluctuation rate during the current batch of waste treatment process, which measures the volatility of physical properties; λ1, λ2, λ3: The adjustment coefficients corresponding to each target, dynamically allocating optimization attention.

[0026] Preferably, the physical property fluctuation is used as a component of the feedback in the strategy feedback optimization module. When the physical property fluctuation of the corresponding link is large, the robustness of the system is maintained through the corresponding adjustment coefficient.

[0027] In summary, the technical effects and advantages of the present invention are as follows: This solution can automatically adjust the operating parameters of the processing equipment based on the different characteristics of the waste (such as hardness, humidity, particle size, etc.), avoiding the problems of over-processing and waste of resources in traditional technologies. In addition, the system of the present invention can dynamically optimize the processing flow on the basis of real-time monitoring of the state of the waste and environmental conditions, so that the waste recycling process is not only more efficient, but also reduces environmental pollution. Through adaptive optimization control and precise classification, the present invention can overcome many defects in traditional waste recycling systems. Compared with manual classification and traditional mechanical sorting, the processing system of the present invention greatly improves the recovery accuracy, reduces manual intervention, and avoids the problem of misclassification. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic diagram of the module structure in the present invention. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0030] like Figure 1 As shown, a construction concrete waste treatment system includes:

[0031] A data acquisition module, which is used to collect physical properties of waste materials to obtain initial data and input the initial data into the model generation and optimization module;

[0032] A model generation and optimization module, which is used to establish a waste material physical property model based on the initial data collected by the data acquisition module, and to update and modify the waste material physical property model through a closed-loop feedback mechanism to output a dynamic waste material physical property model;

[0033] a strategy generation module, the strategy generation module being configured to use the dynamic waste physical property model output by the model generation and optimization module as an environmental state input and output equipment operating parameters of the waste treatment equipment based on the optimization strategy;

[0034] A strategy mapping module maps device operating parameters into physical execution signals through a device instruction mapping model, sends them to corresponding devices, obtains execution results, and inputs the execution results into a strategy feedback optimization module;

[0035] The strategy feedback optimization module identifies the execution deviation of the device operating parameters in the execution result, and feeds back the execution deviation to the strategy generation module to optimize the device operating parameters to obtain the optimized device operating parameters.

[0036] Sensors in waste recycling systems play a crucial role in data acquisition. To accurately perceive the physical characteristics of waste, a series of sensors are deployed to monitor multiple physical properties in real time. These sensors provide a continuous stream of data, allowing the system to adapt to real-time changes.

[0037] The types of data collected include:

[0038] Humidity M t : Reflects the water content of the waste and affects the crushing properties of the waste.

[0039] Particle size D t : Affects the screening effect of waste. The particle size distribution determines the difficulty of screening and sorting.

[0040] Temperature T t :The temperature of the waste during the processing process affects the crushing and screening efficiency.

[0041] Density P t : Determine the quality and texture of waste and assist in subsequent classification and processing.

[0042] These data will be transmitted to the data processing platform through the Internet of Things to form the waste physical property model P f Provide real-time input. Sensors will continuously collect data and transmit it to the data processing system, providing the basis for subsequent dynamic adjustments.

[0043] In the model generation and optimization module, a mathematical model is first established for the multi-dimensional physical properties of the waste (such as particle size, moisture, hardness, and density). Data collected by on-site sensors is then converted into a unified model representation. Different waste properties have different impacts on subsequent treatment processes, so each physical property is assigned a weight. This weighting not only considers the impact of the physical properties themselves on treatment effectiveness but also adjusts based on changes in the waste during the treatment process.

[0044] Waste physical property model P f is represented as follows:

[0045]

[0046] in:

[0047] w i is each physical property X iWeights (such as particle size, humidity, hardness, etc.), w i satisfy

[0048] X i is the value of the i-th physical property of the waste (e.g., particle size D i , humidity M i , hardness H i wait).

[0049] Actual operation:

[0050] Real-time monitoring of waste material humidity M through sensor data t , particle size D t wait.

[0051] For each batch of waste, determine the appropriate weight coefficient w according to the characteristics of the waste i , and dynamically adjust the model.

[0052] Physical modeling provides a precise, multi-dimensional description of each type of waste, ensuring that subsequent processing steps adjust parameters based on the waste's actual characteristics, rather than relying on empirical or fixed processing rules. This flexibility ensures a precise match between waste characteristics and the processing process, laying a solid foundation for intelligent optimization and adaptive control.

[0053] The physical properties of waste materials will change with the environment, weather, and processing process. For example, changes in humidity will affect the difficulty of handling concrete waste, and changes in particle size will affect the crushing efficiency. Therefore, it is necessary to use the real-time data collected by sensors to dynamically update the physical model P f (t), so that it can accurately reflect the current physical characteristics of the waste.

[0054] The physical model of waste is updated after each data collection. The update formula is as follows:

[0055]

[0056] in:

[0057] P f (t) is the physical property model of the waste at time t.

[0058] w i (t) is the weight of each physical property at time t, which is dynamically adjusted according to the changes in sensor data during the waste processing process.

[0059] X i (t) is the i-th physical property value of the waste at time t (e.g., particle size, moisture, etc.).

[0060] Actual operation:

[0061] Collect waste material data such as humidity and particle size in real time.

[0062] Based on these real-time data, the latest state P of the waste characteristic model is calculated. f (t), and constantly adjust the model parameters (weights w i (t)).

[0063] Through this dynamic update process, the system can always automatically adjust the operating strategy in the waste treatment process according to the changes in actual waste.

[0064] By dynamically updating the waste model, changes in waste characteristics can be detected in real time, and weights can be automatically adjusted accordingly, ensuring that the physical model always adapts to environmental changes, improving the accuracy and adaptability of waste processing. For example, in high humidity environments, the humidity weight will automatically increase, and the system will automatically adjust the processing strategy to cope with the impact of high humidity, thereby improving waste recovery rates.

[0065] To ensure the accuracy of the waste physical model, each batch of waste is verified after processing to examine the actual recycling performance and compare it with the physical model. If the treatment effect is not as expected, the weighting and physical characteristics of the waste model are adjusted based on feedback. Specifically, data such as the recycling rate and energy consumption of the waste after processing are used as verification indicators and fed back into the model for adjustment.

[0066] For example, if the crushing efficiency is low during the processing, it may be that the particle size model is inaccurate or the humidity is too high. At this time, the model is adjusted through the feedback mechanism to correct the weight or model parameters. After each treatment, the treatment effect (such as recovery rate) is used as a feedback signal to correct the weight coefficient w of the waste model. i These revised physical property models are applied to the processing of the next batch of waste, forming a closed-loop feedback loop to ensure that the system is always in an optimal state.

[0067] The physical properties of waste materials are constantly changing with environmental changes and waste treatment progress. In order to ensure that the waste recycling system can adapt to these changes at any time, we need to model the physical properties of waste materials P f Perform dynamic updates.

[0068] The model update formula is as follows:

[0069]

[0070] in:

[0071] P f (t): Physical property model of waste at time t;

[0072] w i(t): the weight of the i-th physical characteristic at time t;

[0073] X i (t): The measured value of the ith physical property at time t (e.g., M t ,D t ,T t ,P t );

[0074] λ: regularization coefficient, which controls the effect of the rate of change of physical properties;

[0075] Physical PropertiesX i The rate of change of a characteristic is used to measure the severity of the change.

[0076] Item 1 It is the weighted average of the waste characteristics. The weight w is dynamically adjusted based on the importance of each characteristic to the treatment effect. i (t). The second item It is a regularization term that smooths the changes in the model and prevents the physical characteristics from changing dramatically in a short period of time, which may cause the model to become unstable.

[0077] After each batch of waste is processed, the system will verify and provide feedback on the physical property model based on the actual processing results (such as recovery rate, energy consumption, etc.). If the processing results are not good, the system will automatically correct the weights w of each physical property in the physical model based on the feedback. i (t).

[0078] The feedback weight update formula is as follows:

[0079] w i (t+1)=w i (t)+α·δw i (t)

[0080] in:

[0081] w i (t+1): updated physical property weights;

[0082] w i (t): physical property weight at the current moment;

[0083] α: learning rate, which controls the step size of each adjustment;

[0084] δw i (t): Weight correction based on processing feedback, reflecting the error of indicators such as recovery rate.

[0085] The feedback mechanism continuously optimizes the model through historical processing data and recovery rate indicators, making the model weight w i (t) More consistent with actual waste characteristics.

[0086] The physical properties of waste vary with environmental factors such as season, humidity, and waste type. For example, in a high humidity environment, the system will automatically increase the humidity M t The weight of the subsequent processing strategy increases the crushing intensity or adjusts the size of the sieve hole of the screening machine to ensure that the waste recovery efficiency is not affected by environmental fluctuations. The system can dynamically adjust the processing parameters of the equipment based on real-time data, so that the waste treatment process can adapt to different on-site conditions and avoid the inefficiency caused by fixed strategies. Through real-time data collection and dynamic updating, this step ensures that the waste physical model P f (t) It always matches the actual state of the waste, provides an accurate input basis for subsequent intelligent optimization strategies, and improves the real-time performance, accuracy and resource utilization of the waste treatment system.

[0087] In the strategy generation module, the dynamic waste physical property model P output by the model generation and optimization module is input f (t), the model describes the physical state of the current batch of waste (such as M t Humidity, D t Particle size, P t Density, T t temperature, etc.) is the core input data for the generation of intelligent optimization strategies. The task of this module is to f (t) dynamic state, generating the optimal device control parameters S t , as the input of the subsequent actual processing link. To this end, this solution designs a "physical property driven adaptive strategy generator" (APCOS), whose core mechanism is: f (t) is used as the environmental state input, and the optimization strategy π based on physical characteristics is adopted θ , output a set of operating parameters S of the waste treatment equipment t , such as the speed of the crusher, the mesh size of the screening machine, the wind speed of the sorting equipment, etc. Equipment operating parameters S t Generated through the following smart strategies:

[0088] S t =π θ (P f (t))+β·Ω(P f (t))

[0089] in:

[0090] S t : A collection of equipment control parameters for the waste treatment system (such as crusher speed, screening particle size, sorter wind speed, etc.).

[0091] π θ (P f(t)):Intelligent control strategy generated by APCOS system, based on P f (t) physical characteristics, dynamic output control parameters.

[0092] Ω(P f (t)) is a physical feasibility correction term, which is calculated based on the current physical state of the waste material P. f (t), for π θ The output device parameters are corrected for physical boundaries to ensure that the strategy results conform to physical laws (for example, when the humidity is high, the screening particle size should be increased to prevent clogging).

[0093] β: The adjustment coefficient of the physical correction term, which is used to balance the flexibility and physical rationality of strategy generation.

[0094] Different from the traditional processing parameter setting based on empirical rules, the APCOS system adopts an adaptive strategy π θ , can be updated according to the real-time P f (t) Status (such as M t Increase, D t Particle size distribution changes, etc.), dynamically generate the optimal control parameters of the processing equipment. f (t)), the system embeds physical boundaries related to waste characteristics based on intelligent optimization to prevent strategy generation from exceeding the feasible operating range of the equipment. The APCOS strategy optimization process uses the following "strategy-physics-energy consumption" joint reward mechanism for learning optimization:

[0095]

[0096] in:

[0097] R(S t ): Indicates the adoption of strategy S t The final recycling efficiency index comes from historical processing data feedback.

[0098] C(S t ):Indicates the use of S t The energy consumption index of the rear equipment is used to balance the energy efficiency of the equipment.

[0099] γ: Energy consumption penalty factor, which prevents the system from only pursuing recovery rate and ignoring energy efficiency.

[0100] This joint reward mechanism ensures that the APCOS system generates the control parameter S t When considering the physical properties of the waste P f (t), the balance between recycling efficiency and energy consumption, ensuring that the strategy is not only efficient but also has good energy utilization efficiency. For example, when P f (t) Medium humidity M t Higher, and particle size D tWhen it is large, the system will generate a combination strategy S of "high speed + large sieve hole". t , to avoid problems such as humidity causing screen blockage and large particle size causing low crushing efficiency; on the contrary, when P f (t) Display granularity D t If it is too small, the system will automatically reduce the screening particle size to reduce energy consumption redundancy.

[0101] In the strategy mapping module, the device operation strategy S output by the strategy generation module is input t , which includes the mill speed V t , Screening machine sieve size G t , sorting machine wind speed F t Parameters such as P are based on the physical properties of the waste f (t) is the intelligent optimization result. The task of the strategy mapping module is to convert S t The data is transmitted to the equipment without distortion, realizing the actual physical processing actions such as crushing, screening, and sorting of the waste, completing the landing closed loop from "strategy" to "execution". To this end, the system has designed a standard mapping channel of "strategy → control signal → equipment execution" to ensure S t Can be accurately converted into execution instructions of the device A j (t), the system no longer t Make any modification or re-optimization at the policy level. The device command mapping model is:

[0102] A j (t) = k j ·S t,j

[0103] in:

[0104] A j (t): The control signal of the jth processing equipment (such as the motor speed of the crusher, the mesh size of the screening machine, the wind speed of the sorting machine, etc.).

[0105] k j : The response coefficient of the device, used to convert S t,j The policy parameters in the standardization are mapped into physical execution signals, which directly drive the device actions.

[0106] The system will A j (t) Directly send it to the corresponding device to implement the following physical processing flow:

[0107] Crushing equipment according to V t The waste is crushed to optimize the particle size according to the instructions;

[0108] Screening equipment according to G t , dynamically adjust the sieve size to screen granular materials;

[0109] Sorting equipment according to Ft , adjust the wind speed to complete the separation of recyclable materials and waste residue.

[0110] During execution, the device responds to S in real time t Instructions, complete the preset waste processing action chain, ensure that the crushing, screening and sorting processes are in accordance with S t The optimal working conditions defined by the system are used to avoid any "distortion" or "secondary adjustment" of the strategy. The execution results of each device (such as crushing particle size distribution, screening rate, sorting recovery rate, etc.) and the operating energy consumption data (E j (t)), processing efficiency (Q j (t)) is collected synchronously to form the execution result set R t , output to the strategy feedback optimization module for closed-loop optimization.

[0111] In the policy feedback optimization module, the input is the recovery rate Unit energy consumption E t / Q t 、Equipment operating load L t And material handling stability (such as particle size fluctuation rate D var , humidity fluctuation rate M var ) and other indicators, by designing the "Dynamic Multi-Objective Feedback-Optimization Module (DMOF)", R t With S t Perform coupling analysis to complete the identification strategy S t Under the current physical property conditions P f (t), and based on the deviation, dynamically adjust π in the APCOS policy system (step 3) θ , realizing self-evolutionary optimization of waste treatment systems.

[0112] For construction concrete waste with large fluctuation in physical properties (humidity M t , particle size D t This solution innovatively introduces a "physical property instability suppression regularization term" based on the traditional recovery rate-energy consumption optimization, making the continuous optimization strategy more adaptable to the field. The specific feedback optimization formula is:

[0113]

[0114] in:

[0115] ΔS t : Policy deviation, used to guide π θ Adjustment;

[0116] Recovery rate targets set by the system;

[0117] The recovery rate of actual treatment;

[0118] E t / Q t : unit energy consumption;

[0119] D var 、M var : The particle size fluctuation rate and humidity fluctuation rate during the current batch of waste processing are used to measure the physical property fluctuation;

[0120] λ1, λ2, λ3: adjustment coefficients corresponding to each target, dynamically allocating and optimizing attention.

[0121] In response to the common problems of “large batch differences, drastic fluctuations in humidity and particle size” in concrete waste, we actively take physical property fluctuations as an important part of feedback to prevent the strategy from simply pursuing recovery rate or energy consumption while ignoring the dynamics of material properties, and maintain the robustness of the system. For example: if D var 、M var A higher value indicates that the physical properties of the waste fluctuate greatly during the crushing-screening-sorting process. The system strengthens this effect through λ3, prompting the next round of S t Be more conservative and prevent problems such as unstable equipment load and abnormal energy consumption caused by overly aggressive strategies.

[0122] For strategy π θ To address this issue, we designed a "historical sliding window regularization mechanism" to balance short-term response and long-term convergence, solving the problem of "too fast short-term adjustment and too slow long-term convergence" in traditional systems. The update formula is as follows:

[0123]

[0124] in:

[0125] A new round of optimization strategies;

[0126] Current strategy;

[0127] The average of the strategies in the previous k rounds forms a historical window;

[0128] η: learning rate;

[0129] ξ: Sliding window regularization coefficient, suppressing strategy oscillation.

[0130] Based on the "extreme physical property mutations" that may exist in the waste treatment process (such as the soaring humidity of waste during heavy rain), the system is prevented from fluctuating due to the treatment effect of a single batch, and the strategy quickly converges to the local optimum, thereby reducing the long-term system efficiency.

[0131] That is, the feedback-optimization-strategy update process is:

[0132] Step 4 The device completes execution and collects R t ;

[0133] The DMOF module converts R t With S t Compare and calculate ΔS based on physical property volatility and energy efficiency t ;

[0134] ΔS t As an optimization signal, it is passed to APCOS (step 3) to dynamically update π θ , generate the next round S t ;

[0135] Complete adaptive learning and continuously optimize the system's processing efficiency and energy efficiency.

[0136] The technical solutions in the embodiments of the present application described above have at least the following technical effects or advantages: They can automatically adjust the operating parameters of processing equipment based on the different characteristics of the waste (such as hardness, moisture, particle size, etc.), avoiding the problems of over-processing and resource waste that exist in traditional technologies. Furthermore, the system of the present invention can dynamically optimize the processing flow based on real-time monitoring of the waste state and environmental conditions, making the waste recycling process not only more efficient but also reducing environmental pollution.

[0137] The working principle is as follows: through real-time data collection and feedback, the waste processing process is optimized, and the operating parameters of the processing equipment can be automatically adjusted based on the different characteristics of the waste (such as hardness, humidity, particle size, etc.). On the basis of real-time monitoring of the status of the waste and environmental conditions, the processing process is dynamically optimized, and through adaptive optimization control and precise classification, many defects in the traditional waste recycling system are overcome.

[0138] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A construction concrete waste treatment system, characterized in that: include: A data acquisition module, which is used to collect physical properties of waste materials to obtain initial data and input the initial data into the model generation and optimization module; A model generation and optimization module, which is used to establish a physical property model of the waste material based on the initial data collected by the data acquisition module, and to update and modify the physical property model of the waste material through a closed-loop feedback mechanism to output a dynamic physical property model of the waste material; a strategy generation module, the strategy generation module being configured to use the dynamic waste physical property model output by the model generation and optimization module as an environmental state input and output equipment operating parameters of the waste treatment equipment based on the optimization strategy; A strategy mapping module maps device operating parameters into physical execution signals through a device instruction mapping model, sends them to corresponding devices, obtains execution results, and inputs the execution results into a strategy feedback optimization module; The strategy feedback optimization module identifies the execution deviation of the device operating parameters in the execution result, and feeds back the execution deviation to the strategy generation module to optimize the device operating parameters to obtain the optimized device operating parameters.

2. The construction concrete waste treatment system according to claim 1, characterized in that: The initial data collected by the data collection module include: humidity, particle size, temperature and density of the waste.

3. The construction concrete waste treatment system according to claim 1, characterized in that: The model generation and optimization module establishes a physical property model of the waste by assigning a corresponding weight to each physical property in the initial data. The physical property model of the waste is dynamically updated according to the data collected by the data acquisition module at different times. The weight is adjusted based on the importance of each physical property to the treatment effect, and the dynamic waste physical property model is obtained after updating and correcting.

4. The construction concrete waste treatment system according to claim 1, characterized in that: The physical property model P of the waste in the model generation and optimization module f is represented as follows: Where: w i is each physical property X i The weight, w i satisfy X i is the i-th physical property value of the waste.

5. The construction concrete waste treatment system according to claim 1, characterized in that: The device operating parameters of the strategy generation module are generated by the following strategy: S t =π θ (P f (t))+β·Ω(P f (t)) Among them, S t is the set of equipment control parameters of the material handling system; θ (P f (t)) is based on the physical property model P f (t) physical properties optimization strategy; Ω(P f (t)) is based on the physical property model P f The physical feasibility correction term of the physical characteristics of the waste (t) performs physical boundary correction on the output equipment parameters according to the current physical state of the waste to ensure that the strategy results conform to the physical laws; β is the adjustment coefficient of the physical correction term, which is used to balance the flexibility and physical rationality of strategy generation.

6. The construction concrete waste treatment system according to claim 5, characterized in that: The optimization strategy in the strategy generation module is based on the recycling efficiency index after adopting strategy S and the energy consumption index of the equipment after adopting S, and balances the relationship between the physical properties of the waste, recycling efficiency and energy consumption to complete the optimization.

7. The construction concrete waste treatment system according to claim 1, characterized in that: The policy feedback optimization module uses a historical sliding window regularization mechanism to balance short-term response and long-term convergence in policy updates. Specifically: Where, ΔS t : Policy deviation, used to guide π θ Adjustment; Recovery rate targets set by the system; Actual treatment recovery rate; E t / Q t : Unit energy consumption; D var 、M var : The particle size fluctuation rate and humidity fluctuation rate during the current batch of waste treatment process, which measures the volatility of physical properties; λ1, λ2, λ3: The adjustment coefficients corresponding to each target, dynamically allocating optimization attention.

8. The construction concrete waste treatment system according to claim 7, characterized in that: In the strategy feedback optimization module, the physical property fluctuation is used as a component of the feedback. When the physical property fluctuation of the corresponding link is large, the corresponding adjustment coefficient is used to maintain the robustness of the system.

Images