Multi-agent based construction cost and carbon emission day-by-day simulation method and device, and storage medium

Abstract

The application relates to a multi-agent-based construction cost and carbon emission daily simulation method and device and a storage medium, wherein the method comprises the following steps: determining the basic structures of various agents and modeling the various agents, wherein the types of the agents include worker agents, manager agents, weather agents, building material agents and construction equipment agents, and the basic structures of the agents comprise a static parameter library, a dynamic variable library and an interaction rule library; based on the established various agents, a simulation process is performed, and the carbon emission amount in the simulation process is obtained. Compared with the prior art, the application has the advantages of improving the carbon emission calculation accuracy and the like.

Description

Technical Field

[0001] This invention relates to the field of computer simulation, and in particular to a method, apparatus, and storage medium for daily simulation of construction costs and carbon emissions based on multi-agent systems. Background Technology

[0002] Since the Industrial Revolution, the concentration of carbon dioxide in the atmosphere has increased dramatically, leading to rising temperatures, ocean acidification, and other global environmental problems. The construction industry accounts for as much as 31% of global carbon emissions, and this figure is expected to continue to rise as global construction activities continue. On the other hand, construction costs are a key focus in construction management. Accurately quantifying construction costs and carbon emissions and proposing reasonable optimization measures requires sufficient attention.

[0003] In existing technologies, carbon emission calculations are usually based on the materials and construction period in the construction plan. This method ignores the interaction of various factors and is not accurate. Summary of the Invention

[0004] The purpose of this invention is to provide a method, apparatus, and storage medium for daily simulation of construction costs and carbon emissions based on multi-agent systems.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A daily simulation method for construction cost and carbon emissions based on multi-agent systems includes:

[0007] The basic structure of various intelligent agents is determined and the models of various intelligent agents are constructed. The types of intelligent agents include worker intelligent agents, manager intelligent agents, weather intelligent agents, building material intelligent agents, and construction equipment intelligent agents. The basic structure of the intelligent agents includes a static parameter library, a dynamic variable library, and an interaction rule library.

[0008] Based on the established intelligent agents, the simulation process is executed, and the carbon emissions during the simulation process are obtained.

[0009] Workers of each type of work are set up as independent intelligent agents. The responsibility of the worker intelligent agent is to carry out construction activities under the management of the manager intelligent agent. During the activities, the worker intelligent agents operate construction equipment intelligent agents and consume building material intelligent agents. The static parameters of the worker intelligent agents include the worker's work efficiency in normal state, work efficiency in idle state, work efficiency in overtime state, overtime duration and hourly wage, etc. The dynamic variables of the worker intelligent agents include skill level, daily working hours and idle time. The interaction rule base includes state transition trigger rules, worker skill level improvement rules and worker intelligent agent operation equipment intelligent agent rules.

[0010] The worker agent's state includes four stages: rest, normal work, slack, and overtime. When in rest mode, the worker agent does not engage in any construction activities. Upon reaching the designated working time, the worker agent enters normal work mode. If it receives an overtime instruction from the manager agent, the worker agent enters overtime mode. The worker agent consumes building materials and equipment during normal or overtime modes and enters a slack state periodically. When the worker agent enters a slack state, construction efficiency decreases. When the manager agent conducts on-site inspections, the worker agent immediately ends its slack behavior. After the construction activity is completed, the worker agent returns to rest mode.

[0011] The manager agent's responsibilities include regularly inspecting and supervising the worker agents' construction activities, organizing worker agents to participate in skills training, conducting construction quality inspections to determine whether rework is needed for substandard construction quality, and assessing construction progress to determine whether overtime work is required. The manager agent is only configured with static parameters, including the frequency of on-site inspections during normal working hours, the frequency of on-site inspections during overtime work, and the frequency of organizing worker training. The interaction rule base includes on-site inspection trigger rules, training organization trigger rules, overtime organization trigger rules, and actual progress calculation rules.

[0012] The management agent's state includes three stages: rest, normal work, and overtime. When the management agent is in the rest state, it does not carry out any activities. When the scheduled working time arrives, the management agent determines whether the current construction progress is behind schedule. If it is behind schedule, it issues an overtime instruction to the worker agents. In the normal work and overtime states, the management agent conducts on-site inspections at a set frequency. The management agent organizes skills training activities to improve workers' skill levels and construction efficiency. When the predetermined end time arrives, the management agent updates the current construction progress according to the progress calculation rules and returns to the rest state.

[0013] The weather agent combines daily meteorological data during the simulation process to generate a weather impact coefficient on the worker agent's work efficiency. The specific calculation rules are as follows: when the maximum temperature exceeds 32℃, the worker's work efficiency decreases by 1.3% for every 1℃ increase in temperature; when the wind force is greater than level 4, outdoor high-altitude hoisting, welding and other construction activities are prohibited, and the worker's work efficiency drops to 0; when there is rain or snow, outdoor lifting, installation and other construction activities are prohibited, and the worker's work efficiency drops to 0.

[0014] After each day's simulation, the building material agent receives construction progress information from the manager agent and calculates the material consumption based on the construction progress.

[0015] The construction equipment intelligent agent's state includes three stages: off, running, and standby. When the worker intelligent agent is in a resting state, the construction equipment intelligent agent is in a off state; when the worker intelligent agent starts working, the construction equipment intelligent agent enters a running state; when the worker intelligent agent is in a slack state, the construction equipment intelligent agent enters a standby state; when the worker intelligent agent finishes its work for the day, the construction equipment intelligent agent returns to the off state; after the daily simulation ends, it receives construction progress information from the manager intelligent agent and calculates the equipment consumption based on the construction progress.

[0016] The simulation process, based on the established intelligent agents, is executed to obtain the carbon emissions during the simulation, including:

[0017] Step S1: Based on the duration of the worker agent's work in different working states on the simulation day and the work efficiency in each state, combined with the impact of weather conditions on worker work efficiency, calculate the average construction efficiency of the project for that day:

[0018]

[0019] T t =T work，t +T idle，t +T ot，t

[0020] Where: P t W represents the average construction efficiency of workers on day t. t Let T be the weather impact coefficient on day t. work，t T idle，t T ot，t P represents the cumulative time spent by the worker agent in normal, slack, and overtime states on day t, respectively; work P idle P ot The T values ​​represent the work efficiency of the worker agent in normal working, slack, and overtime states, respectively. t The total working hours on day t;

[0021] Step S2: Calculate the quality pass rate of the completed project:

[0022] Q i，t =R sr，i ·Q sr，i，t +R im，i ·Q im，i，t +R jr，i ·Q jr，i，t

[0023] Among them: Q i，t R represents the average construction quality pass rate of trade i on day t. sr，i R im，i Rjr，i Q represents the proportion of workers in job i with skill levels of advanced, intermediate, and basic. sr，i，t Q im，i，t Q jr，i，t The construction quality pass rate of workers with advanced, intermediate, and basic skill levels in job i, respectively, on day t.

[0024] Step S3: Calculate the actual progress completed each day based on daily construction efficiency, quality pass rate, and working hours.

[0025] AP t =P t ·T t ·Q i，t

[0026] Among them: AP t This represents the actual progress completed on day t.

[0027] Step S4: Calculate material consumption:

[0028] ACM i，t =P t ·T t PCM i，t

[0029] Among them: ACM i，t PCM represents the actual consumption of material i on day t. i，t This represents the planned consumption of material i on day t.

[0030] Step S5: Calculate the equipment's daily operating time consumption:

[0031]

[0032]

[0033] Among them: ACE standby,i，t PCE represents the number of shifts consumed by device i in standby mode on day t. i，t For the planned number of machine shifts consumed by device i on day t, ACE run，i，t Let t be the daily consumption of equipment i in operation.

[0034] Step S6: Calculate the daily construction cost:

[0035] TC t =TCM t +TCE t +TCL t +TC ind，t

[0036] Among them: TC tThe construction cost on day t is TCM. t TCE represents the total material cost on day t. t For the total equipment cost on day t, TCL t Let TC be the total labor cost on day t. ind，t The indirect cost on day t;

[0037] Step S7: Calculate daily construction carbon emissions:

[0038] TE t =TEM t +TEE t

[0039] Among them: TE t For the carbon emissions from construction on day t, TEM t Let TEE be the total carbon emissions generated by the material on day t. t Let be the total carbon emissions generated by the equipment on day t.

[0040] Step S8: Determine whether the construction process is complete. If yes, end the simulation and output the carbon emission time series. Otherwise, return to step S1.

[0041] The total material cost on day t is:

[0042] TCM t =∑ i CM i，t

[0043] CM i，t =ACM i，t ·UP m，i

[0044] Wherein: CM i，t Let UP be the cost of material i on day t. m，i Let i be the unit price of material i;

[0045] The total equipment cost on day t is:

[0046] TCE t =∑ i CE i，t

[0047] CE i，t =PCE i，t ·UP e，i

[0048] Among them: CE i，t For the cost of device i on day t, UP e，i Let i be the unit price of equipment i;

[0049] The total labor cost on day t is:

[0050] TCL t =∑ i CL i，t

[0051] CL i，t =NUM i，t ·[HW i ·(T work，t +T idle，t )+HW ot，i ·T ot，t ]

[0052] Among them: CL i，t NUM represents the wage for job type i paid on day t. i，t HW represents the number of workers of type i used on day t. i The normal hourly wage for job type i; HW ot，i Overtime hourly wage for job type i;

[0053] The indirect cost on day t is:

[0054] TC ind，t =OC t +SF t +TAX t

[0055] OC t =(CL) t +CM t )·r oc

[0056] SF t =(CL) t +CM t +CE t +OC t +PF t )·r sf

[0057] PF t =(CL) t +CM t )·r pf

[0058] TAX t =(CL) t +CM t +CE t +OC t +PF t +SF t )·r tax

[0059] SF t =(CL)t +CM t +CE t +OC t +PF t )·r sf

[0060] Wherein: OC t For the management fee on day t, r oc For the management fee rate, PF t For the profit on day t, r pf For profit margin, SF t For the fee on day t, r sf The rate for regulatory fees, TAX t For the tax on day t, r tax This refers to the tax rate.

[0061] The total carbon emissions generated by the material on day t are:

[0062] TEM t =∑ i CEM i，t

[0063] CEM i，t =ACM i，t ·EF m，i

[0064] Among them: CEM i，t EF represents the carbon emissions generated by material i on day t. m，i Let be the carbon emission factor of material i;

[0065] The total carbon emissions generated by the equipment on day t are:

[0066] TEE t =∑ i CEE i，t

[0067] CEE i，t =ACE run，i，t ·EF e，i ·AE m，i +ACE standby，i，t ·EF e，i ·AE m，i ·α i

[0068] Among them: CEE i，t For the carbon emissions generated by device i on day t, EF e，i For energy i, carbon emission factor, AE m，i α represents the amount of energy consumed by each piece of equipment i per shift. i It is the standby energy coefficient of device i.

[0069] A daily simulation device for construction cost and carbon emissions based on multi-agent systems includes a memory, a processor, and a program stored in the memory, wherein the processor executes the program to implement the method described above.

[0070] A storage medium having a program stored thereon, which, when executed, implements the method described above.

[0071] Compared with existing technologies, the present invention has the following advantages: it can realize the simulation analysis of the construction process and evaluate the construction cost and carbon emission level in days. The invention can also support the adjustment of the attribute values ​​and interaction rules of the intelligent agent, quantify the cost reduction and carbon reduction benefits after adopting various optimization measures, and improve the accuracy of carbon emission measurement. Attached Figure Description

[0072] Figure 1 This is a schematic diagram of the technical route of the present invention;

[0073] Figure 2 A diagram showing the state transitions of different agents;

[0074] Figure 3 Logic diagram for calculating construction costs and carbon emissions. Detailed Implementation

[0075] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0076] A daily simulation method for construction costs and carbon emissions based on multi-agent systems, such as Figure 1 As shown, it includes:

[0077] The basic structure of various intelligent agents is determined and modeling is carried out. The types of intelligent agents include worker intelligent agents, manager intelligent agents, weather intelligent agents, building material intelligent agents and construction equipment intelligent agents. The basic structure of intelligent agents includes a static parameter library, a dynamic variable library and an interaction rule library.

[0078] Construction is the process by which workers, within a specific time and space, follow the instructions of managers and utilize building materials and construction equipment to carry out activities (such as earthwork, steel reinforcement, and concrete work) to transform design drawings into a physical building. The conduct of construction activities is affected by local climate. This invention identifies five types of intelligent agents: worker intelligent agents, manager intelligent agents, weather intelligent agents, building material intelligent agents, and construction equipment intelligent agents. The basic structure of each type of intelligent agent includes a static parameter library, a dynamic variable library, and an interaction rule library. The static parameter library stores inherent attributes that do not change with the specific construction process, such as worker hourly wages, material carbon emission factors, material unit prices, and energy consumption per shift of equipment. The dynamic variable library stores attributes that change during construction, such as worker skill levels and average daily temperature. The interaction rule library stores the state transitions of each intelligent agent and the triggering conditions for various events, as well as the triggering conditions and corresponding actions for interactions between intelligent agents.

[0079] The state transition processes for various intelligent agents are as follows: The worker agent's transition process includes: rest - normal work - slack off - normal work - rest; rest - overtime - slack off - overtime - rest; the manager agent's transition process includes: rest - normal work - rest; rest - overtime - rest; and the equipment agent's transition process is: shutdown - operation - standby - operation - shutdown.

[0080] Worker Agents: Each type of worker is assigned an independent agent. The worker agent's responsibility is to conduct construction activities under the management of the manager agent. These activities involve the operation of equipment agents and the consumption of building material agents. Static parameters of the worker agent include work efficiency under normal conditions, work efficiency under slack conditions, work efficiency under overtime conditions, overtime duration, and hourly wage. Dynamic variables include skill level, daily working hours, and slack duration. The interaction rule base includes state transition trigger rules, worker skill level improvement rules, and rules for worker agents operating equipment agents.

[0081] The manager agent's responsibilities include regularly inspecting and supervising the worker agents' construction activities, organizing worker agents to participate in skills training, conducting construction quality inspections to determine whether rework is necessary for substandard construction quality, and assessing construction progress to determine whether overtime work is required. The manager agent is configured with only static parameters, including the frequency of on-site inspections during normal working hours, the frequency of on-site inspections during overtime, and the frequency of organizing worker training. The interaction rule base includes on-site inspection trigger rules, training organization trigger rules, overtime organization trigger rules, and actual progress calculation rules.

[0082] Weather Agent: The weather agent's responsibility is to store daily meteorological data during construction and calculate the impact of weather on the work efficiency of worker agents. The weather agent only sets dynamic variables including daily average temperature, maximum temperature, minimum temperature, wind force level, rainfall level, and snowfall level; the interaction rule base includes calculation rules for the impact of severe weather on work efficiency.

[0083] Building Material Intelligent Agents: Each type of building material is set up as an independent intelligent agent. The static parameters of the building material intelligent agent mainly include the material's carbon emission factor and unit price; the dynamic variables include the daily planned consumption of building materials based on the construction plan; the interaction rule base includes rules for calculating material consumption and judging material consumption, mainly based on the percentage of construction process progress. The rules for judging material consumption determine the current construction process based on the completed construction progress, and determine the types and quantities of materials planned to be consumed by the relevant process.

[0084] Construction Equipment Intelligent Agents: Each type of construction equipment is set up as an independent intelligent agent. The static parameters of the construction equipment intelligent agent include the equipment's energy consumption intensity, carbon emission factor, energy consumption coefficient in standby mode, and unit price; dynamic variables include the daily planned consumption of construction equipment based on the construction plan; the interaction rule base includes equipment state transition trigger rules and equipment usage calculation and judgment rules. The equipment usage calculation rule is based on the percentage of construction process progress. The equipment usage judgment rule is based on the currently completed construction progress to determine the current construction process, and then determines the relevant process plan to consume the type and number of equipment shifts.

[0085] like Figure 2 As shown, the worker agent's state includes four stages: rest, normal work, slack, and overtime. When in rest mode, the worker agent does not engage in any construction activities. Upon reaching the designated working time, the worker agent enters normal work mode. If it receives an "overtime" instruction from the manager agent, the worker agent enters overtime mode. In normal or overtime mode, the worker agent consumes building materials and equipment, and periodically enters a slack state. When in a slack state, the worker agent engages in activities unrelated to construction, resulting in decreased construction efficiency. When the manager agent conducts on-site inspections, the worker agent immediately ends its slack behavior. After construction activities are completed, the worker agent returns to rest mode.

[0086] The manager's agent status includes three stages: rest, normal work, and overtime. When in rest mode, the manager agent does not engage in any activities. Upon reaching the designated working time, the manager agent determines whether to organize worker agents for overtime based on whether the current construction progress is behind schedule. If overtime is required, the manager agent enters overtime mode; otherwise, it enters normal work mode. During normal and overtime modes, the manager agent conducts on-site inspections at a set frequency. If workers are found to be slacking off, the manager agent corrects this promptly, returning the worker agents to normal work mode. The manager agent also organizes skills training activities to improve worker skills and construction efficiency. Upon reaching the designated end time, the manager agent updates the current construction progress according to the progress calculation rules and returns to rest mode.

[0087] During the simulation, the weather agent combines daily meteorological data to calculate the impact of weather conditions on the work efficiency of the worker agent. The specific calculation rules are as follows: when the maximum temperature exceeds 32℃, the worker's work efficiency decreases by 1.3% for every 1℃ increase in temperature; when the wind force is greater than level 4, outdoor high-altitude hoisting, welding and other construction activities are prohibited, and the worker's work efficiency drops to 0; when there is rain or snow, outdoor lifting, installation and other construction activities are prohibited, and the worker's work efficiency drops to 0.

[0088] After each day's simulation, the material agent receives construction progress information from the manager agent and calculates the material consumption based on the construction progress.

[0089] The equipment intelligence agent's state includes three stages: off, running, and standby. When the worker intelligence agent is in a resting state, the equipment intelligence agent is in the off state; when the worker intelligence agent starts working, the equipment intelligence agent enters the running state; when the worker intelligence agent is in a slack state, the equipment intelligence agent enters the standby state; when the worker intelligence agent finishes its work for the day, the equipment intelligence agent returns to the off state; after the daily simulation ends, it receives construction progress information from the manager intelligence agent and calculates the equipment consumption based on the construction progress.

[0090] Based on the established intelligent agents, the simulation process is executed, and the carbon emissions during the simulation process are obtained, such as... Figure 3 As shown, it includes:

[0091] Step S1: Based on the duration of the worker agent's work in different working states on the simulation day and the work efficiency in each state, combined with the impact of weather conditions on worker work efficiency, calculate the average construction efficiency of the project for that day:

[0092]

[0093] T t =T work,t +Tidle,t +T ot,t

[0094] Where: P t W represents the average construction efficiency of workers on day t. t Let T be the weather impact coefficient on day t. work，t T idle，t T ot，t P represents the cumulative time spent by the worker agent in normal, slack, and overtime states on day t, respectively; work P idle P ot The T values ​​represent the work efficiency of the worker agent in normal working, slack, and overtime states, respectively. t The total working hours on day t;

[0095] Step S2: Calculate the quality pass rate of the completed project. Workers with different skill levels produce building products of varying quality. Only projects that meet the construction quality requirements can be accepted and included in the completed progress. The construction quality pass rate is determined by the production performance of all workers involved in the construction activities; therefore, the quality pass rate is:

[0096] Q i，t =R sr,i ·Q sr,i，t +R im,i ·Q im,i,t +R jr，i ·Q jr，i，t

[0097] Among them: Q i，t R represents the average construction quality pass rate of trade i on day t. sr,i R im,i R jr，i Q represents the proportion of workers in job i with skill levels of advanced, intermediate, and basic. sp,i,t Q im,i,t Q jr，i，t The construction quality pass rate of workers with advanced, intermediate, and basic skill levels in job i, respectively, on day t.

[0098] Step S3: Calculate the actual progress completed each day based on daily construction efficiency, quality pass rate, and working hours.

[0099] AP t =P t ·T t ·Q i，t

[0100] Among them: AP t This represents the actual progress completed on day t.

[0101] Step S4: Calculate material consumption. The type and quantity of materials consumed daily are calculated based on the actual workload of the day, determined by the construction efficiency and working hours. Therefore, the material consumption is:

[0102] ACM i，t =P t ·T t PCM i，t

[0103] Among them: ACM i，t PCM represents the actual consumption of material i on day t. i，t This represents the planned consumption of material i on day t.

[0104] Step S5: Calculate the daily equipment consumption per shift. The type and number of equipment shifts consumed each day are calculated based on the actual workload of the day, determined by the construction efficiency and working hours. The daily equipment consumption is calculated separately for both running and standby modes. Therefore, the daily equipment consumption per shift is:

[0105]

[0106]

[0107] Among them: ACE standby,i，t PCE represents the number of shifts consumed by device i in standby mode on day t. i，t For the planned number of machine shifts consumed by device i on day t, ACE run，i，t Let t be the daily consumption of equipment i in operation.

[0108] Step S6: Calculate the daily construction cost, which includes direct and indirect costs. Direct costs include material, equipment, and labor costs, while indirect costs include management fees, regulatory fees, and taxes. These costs are linearly related to direct costs. Therefore, the construction cost is:

[0109] TC t =TCM t +TCE t +TCL t +TC ind，t

[0110] Among them: TC t The construction cost on day t is TCM. t TCE represents the total material cost on day t. t For the total equipment cost on day t, TCL t Let TC be the total labor cost on day t. ind，t The indirect cost on day t;

[0111] The total material cost on day t is:

[0112] TCMt=∑ i CM i，t

[0113] CM i，t =ACM i，t ·UP m，i

[0114] Wherein: CM i，t Let UP be the cost of material i on day t. m，i Let i be the unit price of material i;

[0115] The total equipment cost on day t is:

[0116] TCE t =∑ i CE i，t

[0117] CE i，t =PCE i，t ·UP e，i

[0118] Among them: CE i，t For the cost of device i on day t, UP e，i Let i be the unit price of equipment i;

[0119] The total labor cost on day t is:

[0120] TCL t =∑ i CL i,t

[0121] CL i，t =NUM i，t ·[HW i ·(T work，t +T idle，t )+HW ot,i ·T ot，t ]

[0122] Among them: CL i，t NUM represents the wage for job type i paid on day t. i，t HW represents the number of workers of type i used on day t. i The normal hourly wage for job type i; HW ot，i Overtime hourly wage for job type i;

[0123] The indirect costs on day t are:

[0124] TC ind，t =OC t +SF t +TAX t

[0125] OC t =(CL) t +CM t )·r oc

[0126] SF t =(CL) t +CM t +CE t +OC t +PF t )·r sf

[0127] PF t =(CL) t +CM t )·r pf

[0128] TAX t =(CL) t +CM t +CE t +OC t +PF t +SF t )·r tax

[0129] SF t =(CL) t +CM t +CE t +OC t +PF t )·r sf

[0130] Wherein: OC t For the management fee on day t, r oc For the management fee rate, PF t For the profit on day t, r pf For profit margin, SF t For the fee on day t, r sf The rate for regulatory fees, TAX t For the tax on day t, r tax This refers to the tax rate.

[0131] Step S7: Calculate daily construction carbon emissions. Construction carbon emissions include carbon emissions from the consumption of building materials and carbon emissions from the consumption of energy by construction equipment. Both are calculated using the carbon emission factor method. Therefore, the construction carbon emissions are:

[0132] TE t =TEM t +TEE t

[0133] Among them: TE t For the carbon emissions from construction on day t, TEM t Let TEE be the total carbon emissions generated by the material on day t. t Let be the total carbon emissions generated by the equipment on day t.

[0134] The total carbon emissions generated by the materials on day t are:

[0135] TEM t =∑ i CEM i，t

[0136] CEM i，t =ACM i，t ·EF m，i

[0137] Among them: CEM i，t EF represents the carbon emissions generated by material i on day t. m，i Let be the carbon emission factor of material i;

[0138] The total carbon emissions generated by the equipment on day t are:

[0139] TEE t =∑ i CEE i，t

[0140] CEE i，t =ACE run，i，t ·EF e，i ·AE m，i +ACE standby,i，t ·EF e，i ·AE m，i ·α i

[0141] Among them: CEE i，t For the carbon emissions generated by device i on day t, EF e，i For energy i, carbon emission factor, AE m,i α represents the amount of energy consumed by each piece of equipment i per shift. i It is the standby energy coefficient of device i.

[0142] Step S8: Determine whether the construction process is complete. If yes, end the simulation and output the carbon emission time series. Otherwise, return to step S1.

[0143] The following is a multi-agent simulation model created using the main construction process of a steel structure factory as an example. This simulation is implemented based on the Java language and the Anylogic platform. Worker agents such as senior installers, junior installers, senior painters, and junior painters were established; a manager agent; a weather agent; material agents such as steel columns, steel beams, welding rods, color steel floor slabs, wall frames, and fireproof coatings; and equipment agents such as truck cranes, AC arc welding machines, abrasive wheel cutting machines, electric hammers, and electric air compressors. Relevant data was obtained through on-site investigations and literature reviews. The agents were initialized and assigned values, simulations were conducted, and optimization measures were proposed based on the results.

[0144] It consists of three steps:

[0145] 1. Intelligent agent interaction and dynamic variable updates

[0146] Based on a multi-agent simulation model, the main construction process of the factory building is simulated with a time step of one day. Within each time step, each agent interacts according to the set interaction rules and follows the calculation rules to call static parameters to update the values ​​of dynamic variables.

[0147] 2. Calculate the cost and construction carbon emissions and output the results to an Excel spreadsheet.

[0148] Based on static parameters and updated dynamic variables, the calculation formulas for construction efficiency, construction quality pass rate, completed progress, actual material consumption, actual equipment consumption, direct cost, indirect cost, material carbon emissions, and equipment carbon emissions are called in sequence, and the daily simulation results are output to an Excel spreadsheet.

[0149] 3. Evaluation and optimization of simulation results

[0150] After obtaining the simulation results, construction units and contractors can propose optimization measures based on the simulation results, and rerun the simulation model by changing parameter values, interaction rules, etc., to simulate the construction situation after the optimization measures are taken and calculate the cost and carbon emission results.

[0151] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A daily simulation method for construction cost and carbon emissions based on multi-agent systems, characterized in that, include: The basic structure of various intelligent agents is determined and the models of various intelligent agents are constructed. The types of intelligent agents include worker intelligent agents, manager intelligent agents, weather intelligent agents, building material intelligent agents, and construction equipment intelligent agents. The basic structure of the intelligent agents includes a static parameter library, a dynamic variable library, and an interaction rule library. Based on the established intelligent agents, the simulation process is executed, and the carbon emissions during the simulation process are obtained; Workers of each type of work are set up as independent intelligent agents. The responsibility of the worker intelligent agent is to carry out construction activities under the management of the manager intelligent agent. During the activities, the worker intelligent agents operate construction equipment intelligent agents and consume building material intelligent agents. The static parameters of the worker intelligent agents include the worker's work efficiency in normal state, work efficiency in idle state, work efficiency in overtime state, overtime duration, and hourly wage. The dynamic variables of the worker intelligent agents include skill level, daily working hours, and idle time. The interaction rule base includes state transition trigger rules, worker skill level improvement rules, and worker intelligent agent operation equipment intelligent agent rules. The simulation process, based on the established intelligent agents, is executed to obtain the carbon emissions during the simulation, including: Average construction efficiency calculation steps: Based on the duration of the worker agent in different working states on the simulation day and the work efficiency in each state, combined with the impact of weather conditions on worker work efficiency, the average construction efficiency of the project on that day is calculated comprehensively. in: For workers Average daily construction efficiency For the first Weather impact coefficient of the day The first The cumulative time that the Tiangong AI agent is in normal, slack, or overtime states; These represent the work efficiency of the worker agent when it is in normal working, slacking off, or working overtime, respectively. For the first Total working hours per day; Calculate material consumption: in: For the first Tian Shi Materials The actual consumption, For the first Tian Shi Materials The planned consumption; Calculate the machine-hour consumption of the equipment: in: For the first Tian Shi Equipment The number of shifts consumed in standby mode For the first Tian Shi Equipment The planned consumption of time slots. For the first Tian Shi Equipment Daily consumption during operation; The first t The total carbon emissions generated by Tianshi Materials are in: For the first t Tian Shi Materials j Carbon emissions generated For materials j Carbon emission factors; The first t The total carbon emissions generated by the Tian Shi equipment are: in: For the first Tian Shi Equipment Carbon emissions generated For energy carbon emission factors, For each shift of equipment The amount of energy consumed, It is equipment The energy consumption coefficient for waiting to work.

2. The method for daily simulation of construction cost and carbon emissions based on multi-agent systems according to claim 1, characterized in that, The worker agent's state includes four stages: rest, normal work, slack, and overtime. When in rest mode, the worker agent does not engage in any construction activities. Upon reaching the designated working time, the worker agent enters normal work mode. If it receives an overtime instruction from the manager agent, the worker agent enters overtime mode. The worker agent consumes building materials and equipment during normal or overtime modes and enters a slack state periodically. When the worker agent enters a slack state, construction efficiency decreases. When the manager agent conducts on-site inspections, the worker agent immediately ends its slack behavior. After the construction activity is completed, the worker agent returns to rest mode. The manager agent's responsibilities include regularly inspecting and supervising the worker agents' construction activities, organizing worker agents to participate in skills training, conducting construction quality inspections to determine whether rework is needed for substandard construction quality, and assessing construction progress to determine whether overtime work is required. The manager agent is only configured with static parameters, including the frequency of on-site inspections during normal working hours, the frequency of on-site inspections during overtime work, and the frequency of organizing worker training. The interaction rule base includes on-site inspection trigger rules, training organization trigger rules, overtime organization trigger rules, and actual progress calculation rules. The manager's intelligent agent states include three stages: rest, normal work, and overtime. When the manager's intelligent agent is in the rest state, it does not carry out any activities. When the scheduled working time arrives, the manager's intelligent agent determines whether the current construction progress is behind schedule. If it is behind schedule, it issues an overtime instruction to the worker intelligent agents. In the normal work and overtime states, the manager's intelligent agent conducts on-site inspections at a set frequency. The manager's intelligent agent organizes skills training activities to improve the workers' skill levels and construction efficiency. When the predetermined end time arrives, the manager's intelligent agent updates the current construction progress according to the progress calculation rules and returns to the rest state.

3. The daily simulation method for construction cost and carbon emissions based on multi-agent systems according to claim 2, characterized in that, The weather agent combines daily meteorological data during the simulation process to generate a weather impact coefficient on the worker agent's work efficiency. The specific calculation rules are as follows: when the maximum temperature exceeds 32℃, the worker's work efficiency decreases by 1.3% for every 1℃ increase in temperature; when the wind force is greater than level 4, outdoor high-altitude hoisting, welding and other construction activities are prohibited, and the worker's work efficiency drops to 0; when there is rain or snow, outdoor lifting, installation and other construction activities are prohibited, and the worker's work efficiency drops to 0.

4. The daily simulation method for construction cost and carbon emissions based on multi-agent systems according to claim 3, characterized in that, After each day's simulation ends, the material agent receives construction progress information from the manager agent and calculates the material consumption based on the construction progress.

5. The daily simulation method for construction cost and carbon emissions based on multi-agent systems according to claim 4, characterized in that, The construction equipment intelligent agent's state includes three stages: off, running, and standby. When the worker intelligent agent is in a resting state, the construction equipment intelligent agent is in a off state; when the worker intelligent agent starts working, the construction equipment intelligent agent enters a running state; when the worker intelligent agent is in a slack state, the construction equipment intelligent agent enters a standby state; when the worker intelligent agent finishes its work for the day, the construction equipment intelligent agent returns to the off state; after the daily simulation ends, it receives construction progress information from the manager intelligent agent and calculates the equipment consumption based on the construction progress.

6. The daily simulation method for construction cost and carbon emissions based on multi-agent systems according to claim 5, characterized in that, The process of executing the simulation based on various established intelligent agents and obtaining the carbon emissions during the simulation also includes: Calculate the quality pass rate of completed projects: in: For the first Timing, Job The average construction quality pass rate , Each type of work The proportion of workers with skill levels at the advanced, intermediate, and basic levels. , , The first Timing, Job The pass rate of construction quality for workers with skill levels of advanced, intermediate, and basic. The actual progress completed each day is calculated based on daily construction efficiency, quality pass rate, and working hours: in: For the first The actual progress completed in one day; Calculate daily construction costs: in: The construction cost on day t is... For the first t Total material cost per day For the first t Total equipment cost per day For the first t Total daily labor costs For the first t Indirect costs per day; Calculate daily construction carbon emissions: in: For the first t Construction carbon emissions at opportune times For the first t Total carbon emissions from Tian Shi materials For the first t Total carbon emissions generated by Tian Shi equipment; Determine whether the construction process is complete. If yes, end the simulation and output the carbon emission time series; otherwise, return to the average construction efficiency calculation step.

7. The method for daily simulation of construction cost and carbon emissions based on multi-agent systems according to claim 6, characterized in that, The first t The total material cost for the day is: in: For the first t Tian Shi Materials j The cost, For materials j The unit price; The first t The total equipment cost for the day is: in: For the first t Tian Shi Equipment k The cost, k For equipment k The unit price; The first t The total labor cost per day is: in: For the first Jobs paid daily wages, For the first The type of work used by the time The number of people, The normal hourly wage for job type i; For job type Overtime hourly wage; The first t The indirect costs per day are: = + + + ) + + + ) + ) + + + ) + + + ) in: For the first Management fee per day The management fee rate, For the first Daily profit For profit margin, For the first The daily fee, The fee rate is the fee rate for regulatory fees. For the first Taxes per day This refers to the tax rate.

8. A daily simulation device for construction cost and carbon emissions based on multi-agent systems, comprising a memory, a processor, and a program stored in the memory, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1-7.

9. A storage medium having a program stored thereon, characterized in that, When the program is executed, it implements the method as described in any one of claims 1-7.

Images