CN104715288A - Nonlinear agricultural non-point source pollution control method - Google Patents

Abstract

The invention discloses a nonlinear agricultural non-point source pollution control method. The method comprises the steps that data are collected, a watershed hydrologic model is established, verification is conducted, an interactive algorithm is adopted in a nonlinear non-point source pollution control model, a submodel is split, an upper bound submodel and a lower bound submodel are subjected to linearized solution, 11 risk situations are defined by changing the risk level rho, a linear equation is solved to acquire the agricultural optimal planting area (please see the specifications for the formula) under the 11 risk situations and the optimal dumping right (please see the specifications for the formula) obtained through transaction in each agricultural area, and the comprehensive degree of satisfaction (please see the specifications for the formula) of environment pollution under the 11 risk situations is further acquired. According to the method, the indefinite information characteristics of random numbers, interval numbers, fuzzy numbers and the like are comprehensively considered, the production plan can be helped to be adjusted for a researched river basin, and emission is made to meet the environment requirement.

Description

A kind of non-linear agricultural nonpoint source pollution control method

Technical field

The invention belongs to water quality management field, be specifically related to a kind of non-linear agricultural nonpoint source pollution control method.

Background technology

Agricultural is the mainstay industry of national economy.Along with population increases, increasing to the demand of grain, while agricultural output increases fast, environmental problem is also following.Agricultural nonpoint source pollution load is large, and pollute wide, difficulty of governance is large.For improving grain yield, a large amount of fertilizer and pesticide is applied in farmland.According to statistics, in 2012, state-owned 57,000,000 tons of chemical fertilizer are devoted in farmland, under rainfall and action of topography, these materials are very easily along with rainwash and soil erosion flow into water body, serious water ecology is caused to destroy, such as body eutrophication, downstream ecology safety and people ' s health happiness in serious threat.In addition, non-point pollution is in vast rural area, and infrastructure and the standard system of environmental protection extremely lack, and is badly in need of setting up scientific and reasonable non-point source pollution control and administrative mechanism, manages water quality from source.Emission Trading System is as a kind of effective means of water quality management, can administrative division be crossed over, reasonable distribution dumping right between each farming region, thus realize distributing rationally of environmental resource, under the prerequisite ensureing agricultural output and economic benefit, reduce agricultural nonpoint source pollution on the whole.

In water quality management system, there is a large amount of complicacy and uncertainty, such as, because rainfall and rainwash have randomness, non-point source pollution load also has random character; Agricultural, economy and technical data are incomplete, and statistics also exists error; There is observational error in hydrologic observation data; Basin attribute data has time heterogeneity etc. above and spatially.In addition, the mobility of random data also can cause correlation parameter to change near average, causes systematic jitters.Within multiple planning stage, above-mentioned uncertain reciprocation, causes large amount of complex sex chromosome mosaicism.At present, in the uncertainty management of agriculture water quality management system and the research application aspect of decision-making, carry out a large amount of exploration work, but still had some limitations.Such as, not enough to the research of condition of uncertainty down blow power Trading System, lack the quantitative test to Emission trading system risk, thus it is actual to be difficult to reflection, propose scientific and reasonable water quality management and Emission trading scheme, these problems become the bottleneck of serious restriction water quality management day by day.

Summary of the invention

In order to overcome above-mentioned defect, the object of the present invention is to provide a kind of non-linear agricultural nonpoint source pollution control method.

A kind of non-linear agricultural nonpoint source pollution control method, comprises the steps:

S1. collect and arrange basin natural data and social data, in conjunction with two stage stochastic programming method, Interval Programming method, random robust planing method and system ambiguous planing method, build Interval Fuzzy two benches robust Model for Multi-Objective Optimization and Watershed Hydrologic Models, and water basin combination literary composition model, simulation flow anomaly physical mechanism, prediction rainwash;

S2. set up hydrological model is verified, based on the hydrological model after checking, carry out hydrologic(al) prognosis modeling effort;

S3. based on base flow separation technology, from channel flow process, extract rainwash data, obtain overland flow hydrograph;

S4. based on Monte Carlo technique matching rainwash probability cumulative distribution function, pair distribution function carries out sliding-model control, obtains the rainwash discrete data under multiple probability level, sets up objective function and constraint condition, builds and solving-optimizing model;

S5. based on rainwash discrete data estimation non-point source load, by result of calculation unbalanced input non-point source pollution control model, docking of analogue technique and Optimized model is realized;

S6. interactive remote teaching is adopted to non-linear non-point source pollution control model, split submodel, respectively linearization is carried out to upper bound submodel and lower bound submodel and solve;

S7. by changing risk level ρ, define 11 kinds of risky situations, under 11 kinds of risky situations, risk level value is respectively 0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.Risk level represents the attention degree of supvr to system risk, risk level is higher, show that supvr more payes attention to system risk, such as, ρ=1 shows that supvr considers the impact of system risk on system in earnest, by system risk value based on the thought of risk averse in non-point source Emission trading planning process incorporate in decision process; ρ=0 shows that supvr gives no thought to system risk based on the thought of risk-neutral in non-point source Emission trading planning process, planning risk value be defaulted as 0; ρ=0.5 shows that supvr considers system risk in non-point source Emission trading process, and by the real system value-at-risk of 0.5 times, namely include in the calculating of decision process.

S8. by solving linear equation, agriculture optimum cultivated area under these 11 kinds of risky situations is obtained and each farming region is by handing over facile optimum dumping right and the comprehensive satisfaction of environmental pollution under obtaining 11 kinds of risky situations

Further, natural data described in step S1 and social data comprise watershed unit data, river network of watershed data, Soil attribute data, for many years weather data, agrotype, annual production, agricultural product price, cultivated area and comprise the agricultural planting cost of agricultural chemicals, chemical fertilizer, charges for water and electricity.

Further, Watershed Hydrologic Models described in step S1 is Watershed-scale distributed hydrological model, and the algorithm of employing is:

${\mathrm{SW}}_t={\mathrm{SW}}_0+\underset{i=1}{\overset{t}{\mathrm{\Σ}}}(R_{\mathrm{day}}-Q_{\mathrm{surf}}-E_a-w_{\mathrm{seep}}-Q_{\mathrm{gw}})$

$Q_{\mathrm{surf}}=\frac{{(R_{\mathrm{day}}-I_a)}^2}{(R_{\mathrm{day}}-I_a+S)}$

$S=25.4(\frac{1000}{\mathrm{CN}}-10)$

Wherein, the meaning of parameters that represents of each symbol is as follows:

SW _tsoil moisture content (mm H ₂o);

SW ₀initial soil moisture content (mm H ₂o);

R _daydaily rainfall (mm H ₂o);

Q _surfa day flow path surface (mm H ₂o);

E _adaily evaporation amount (mm H ₂o);

W _seep(the mm H of milliosmolarity under the holard ₂o);

Q _gwunderground water diurnal courses amount (mm H ₂o);

I _arainfall spurt value (mm H ₂o);

S is hysteresis index;

CN is runoff curve number;

I is time (day);

T is duration (day).

Further, described step S4 builds and in solving-optimizing model; By systematic uncertainty with the form body of random number and interval number now in this process.

Further, to set up respectively in described step S7 under Trading System and not non-point source pollution control model under Trading System, as follows:

Under A, Trading System:

Objective function:

(1). system Maximum Satisfaction

Maxλ ^±

(2). total nitrogen total phosphorus discharge capacity is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E({\tilde{Y}}_{\mathrm{Nit}}+{\tilde{Y}}_{\mathrm{Pit}})$

(3). total nitrogen total phosphorus discharge capacity undulatory property is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}+{\widetilde{\mathrm{DE}}}_{\mathrm{Pit}})\right]$

Meet constraint condition:

$\begin{array}{c}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}{B}_{\mathrm{it}}^{\±}(S_{\mathrm{it}}^{-}+\mathrm{\α\Δ}{S}_{\mathrm{it}})-\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}+{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}})\right]\\ -\mathrm{\ρ}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}+{\widetilde{\mathrm{DE}}}_{\mathrm{Pit}})\right]\≥{\mathrm{\Ω}}_{\mathrm{opt}}^{-}+{\mathrm{\λ}}^{\±}({\mathrm{\Ω}}_{\mathrm{opt}}^{+}-{\mathrm{\Ω}}_{\mathrm{opt}}^{-})\end{array}$

$\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({\tilde{Q}}_N{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Nit}})\≤T_{\mathrm{Nt}}^{\±},\∀t$

$\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({\tilde{Q}}_P^{\±}{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Pit}})\≤T_{\mathrm{Pt}}^{\±},\∀t$

$0\≤S_{\mathrm{it}}^{\±}\≤S_{\mathrm{it}\max },\∀i,t$

${\tilde{Y}}_{\mathrm{Nit}}\≤{\tilde{Q}}_N{x}_{\mathrm{it}}^{\±},\∀i,t$

${\tilde{Y}}_{\mathrm{Pit}}\≤{\tilde{Q}}_P{x}_{\mathrm{it}}^{\±},\∀i,t$

${\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}\≥{\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}-E\left[{\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}\right],\∀i,t$

${\widetilde{\mathrm{DE}}}_{\mathrm{Pit}}\≥{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}}-E\left[{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}}\right],\∀i,t$

0≤λ ^±≤1

Under B, nontransaction system:

Objective function:

(1). system Maximum Satisfaction

Maxλ ^±

(2). total nitrogen total phosphorus discharge capacity is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E({\tilde{Y}}_{\mathrm{Nit}}+{\tilde{Y}}_{\mathrm{Pit}})$

(3). total nitrogen total phosphorus discharge capacity undulatory property is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}+{\widetilde{\mathrm{DE}}}_{\mathrm{Pit}})\right]$

Meet constraint condition:

$\begin{array}{c}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}{B}_{\mathrm{it}}^{\±}(S_{\mathrm{it}}^{-}+\mathrm{\α\Δ}{S}_{\mathrm{it}})-\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}+{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}})\right]\\ -\mathrm{\ρ}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}+{\widetilde{\mathrm{DE}}}_{\mathrm{Pit}})\right]\≥{\mathrm{\Ω}}_{\mathrm{opt}}^{-}+{\mathrm{\λ}}^{\±}({\mathrm{\Ω}}_{\mathrm{opt}}^{+}-{\mathrm{\Ω}}_{\mathrm{opt}}^{-})\end{array}$

${\tilde{Q}}_N{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Nit}}\≤(\mathrm{\α}\frac{P{Q}_i}{\underset{i}{\mathrm{\Σ}}P{Q}_i}+\mathrm{\β}\frac{{\mathrm{GDP}}_i}{\underset{i}{\mathrm{\Σ}}{\mathrm{GDP}}_i}+\mathrm{\χ}\frac{Q_i}{\underset{i}{\mathrm{\Σ}}{Q}_i})T_{\mathrm{Nt}}^{\±},\∀t$

${\tilde{Q}}_P{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Pit}}\≤(\mathrm{\α}\frac{P{Q}_i}{\underset{i}{\mathrm{\Σ}}P{Q}_i}+\mathrm{\β}\frac{{\mathrm{GDP}}_i}{\underset{i}{\mathrm{\Σ}}{\mathrm{GDP}}_i}+\mathrm{\χ}\frac{Q_i}{\underset{i}{\mathrm{\Σ}}{Q}_i})T_{\mathrm{Pt}}^{\±},\∀t$

$0\≤S_{\mathrm{it}}^{\±}\≤S_{\mathrm{it}\max },\∀i,t$

${\tilde{Y}}_{\mathrm{Nit}}\≤{\tilde{Q}}_N{S}_{\mathrm{it}}^{\±},\∀i,t$

${\tilde{Y}}_{\mathrm{Pit}}\≤{\tilde{Q}}_P{x}_{\mathrm{it}}^{\±},\∀i,t$

${\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}\≥{\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}-E\left[{\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}\right],\∀i,t$

${\widetilde{\mathrm{DE}}}_{\mathrm{Pit}}\≥{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}}-E\left[{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}}\right],\∀i,t$

0≤λ ^±≤1

Wherein, the meaning of parameters that represents of each symbol is as follows:

+ and-represent the upper and lower bound of interval parameter respectively;

~ represent non-linear stochastic parameter;

I refers to farming region; T is project period; J is probability level;

Ω _optit is the aspiration level of system net proceeds;

λ is decision maker about the comprehensive satisfaction of system benefit and environmental pollution;

B _itthe unit area land revenue (RMB$) of i farming region when t project period;

S _itthe land area (ha) of i farming region when t project period;

π is decision-making coefficient;

ρ is the risk level under kinds of risks sight;

EP _niand EP _pithe fine (RMB$) exceeded standard suffered by 1kg blowdown respectively;

Y _nitand Y _pittotal nitrogen and total phosphorus discharge beyond standards amount (kg) respectively;

DE _nijtand DE _pijtthat i region discharge beyond standards total nitrogen and total phosphorus receive environment fine and the deviation (RMB$) of imposing a fine average level respectively;

Q _nijand Q _pijthe unit area blowdown flow rate (kg) of i farming region under j probability level respectively;

T _ntand T _ptthe total emission volumn standard (kg) about total nitrogen total phosphorus in basin respectively;

PO _iit is the size of population (people) of i farming region;

GDP _iit is the total output value of i farming region;

α, β, χ are the weight of population, total output value, blowdown flow rate in dumping right assigning process respectively.

Further, said optimum dumping right in described step S8, its mechanism of exchange is by model calculation, obtains the dumping right of each farming region reallocation under agriculture Emission trading mechanism, and the difference of getting both obtains each farming region under transaction framework and specifically buys and sells situation.

Beneficial effect of the present invention: emphasis carries out the research of uncertain non-point source Emission trading best mechanism, hydrological simulation technology is combined with interval two benches robust Optimal methods, various uncertainty and complicacy in overall analysis system, and combine the principle of avoiding risk, control water pollution by agricultural non-point source Emission Trading System.The application of non-linear agricultural nonpoint source pollution Controlling model in the identification of Emission trading best mechanism and process of establishing, for establishing and improve agricultural nonpoint source pollution control system and mechanism provides science support, promote agricultural sciences, sustainable development, benefit vast farmers.

Accompanying drawing explanation

Fig. 1 non-linear agricultural nonpoint source pollution controlling planning model system operation frame figure;

Fig. 2 fragrant small stream river valley rainwash forecasting process line;

Optimum cultivated area under Fig. 3 transaction and not mechanism of exchange;

Fig. 4 total nitrogen Emission trading best mechanism.

Embodiment

Below in conjunction with the drawings and specific embodiments, the present invention will be further described.

The fragrant small stream river valley that the present invention chooses is positioned at Hubei China and economizes, and geographic coordinate is between 30 ° of 57 ' N, and 31 ° of 34 ' N and 10 ° of 25 ' E, between 111 ° of 06 ' E, comprises Zhao Jun town, mouth of a gorge town, Gu Fu town, Gui Zhou town and Nanyang town.Xiang Xi rises in Shennongjia in river, drainage area 3099km ², have Gu Fuhe, Gao Lanhe, nine rush tributary, three, river.Phosphorus ore is rich in this basin, and be Chinese the third-largest Chan Lin district, fragrant small stream river valley belongs to subtropics monsoon continental climate, and rainfall is abundant, and May is the pluvial period to September.Farmland flow point cloth along the river in basin, In The Soils is yellowish soil, rendzinas and purple soil.Chief crop has paddy rice, wheat, corn, potato, oranges and tangerines and tealeaves.

Fragrant small stream river valley is positioned at mountain farming area, and topographic relief is large, and the region that the gradient is greater than 25 ° accounts for more than 51.2%, and this causes soil erosion extremely serious in rainy season, and in farmland, a large number of nutrients flows into Xiang Xi river thereupon.According to statistics, Xiang Xi river receives nearly 1112 tons of total nitrogens and 304 tons of total phosphorus every year, but extremely lack for the control measures of non-point pollution, this exacerbates Xiang Xi river Status of Non-point Source Pollution further, and the healthy happiness of river ecological safety and littoral resident in serious threat.Therefore, in the urgent need to setting up scientific and reasonable fragrant small stream river valley agricultural non-point source Emission trading mechanism, controlling non-point pollution, recovering environmental health, maintain fragrant small stream river valley agricultural sustainable development.

Room calculating simulation by experiment, is analyzed the application under inventive method under Trading System and not Trading System.

One, Watershed-scale distributed hydrological model is set up

Hydrological distribution model is selected to carry out simulation and forecast to fragrant small stream river valley each farming region rainwash.The main advantage of hydrological distribution model is its physical mechanism that fully can reflect River Basin Hydrology process; Can combine closely with GIS technology, the special heterogeneity of process distributed information and data; There is higher counting yield and provide many algorithms to select according to actual needs for user.Analog result is as shown in Fig. 2 fragrant small stream river valley rainwash forecasting process line.The algorithm that described Watershed-scale distributed hydrological model adopts is:

${\mathrm{SW}}_t={\mathrm{SW}}_0+\underset{i=1}{\overset{t}{\mathrm{\Σ}}}(R_{\mathrm{day}}-Q_{\mathrm{surf}}-E_a-w_{\mathrm{seep}}-Q_{\mathrm{gw}})$

$Q_{\mathrm{surf}}=\frac{{(R_{\mathrm{day}}-I_a)}^2}{(R_{\mathrm{day}}-I_a+S)}$

$S=25.4(\frac{1000}{\mathrm{CN}}-10)$

Wherein, the meaning of parameters that represents of each symbol is as follows:

SW _tsoil moisture content (mm H ₂o);

SW ₀initial soil moisture content (mm H ₂o);

R _daydaily rainfall (mm H ₂o);

Q _surfa day flow path surface (mm H ₂o);

E _adaily evaporation amount (mm H ₂o);

W _seep(the mm H of milliosmolarity under the holard ₂o);

Q _gwunderground water diurnal courses amount (mm H ₂o);

I _arainfall spurt value (mm H ₂o);

S is hysteresis index;

CN is runoff curve number;

I is time (day);

T is duration (day).

Two, Uncertainty Management

Model of the present invention by hydrological simulation technology is combined with Interval Fuzzy two benches robust Multipurpose Optimal Method, a large amount of uncertainty existed in the non-point source Emission trading system of overall treatment Xiang Xi river and complicacy.First, by hydrological distribution model, take into full account the temporal-spatial heterogeneity of each hydrographic features in rainwash runoff process, as the change in time and space of meteorological attribute, soil attribute; Secondly, take into full account a large amount of stochastic uncertainty, bounded-but-unknown uncertainty and fuzzy uncertainty existed in Emission trading system, by systematic uncertainty with the form of random number and interval number, in the structure being embodied in Optimized model and solution procedure; Finally consider system risk to evade, the system risk value random fluctuation due to economic punishment caused puts into Optimized model objective function.

Three, to determine under Trading System and the not optimum cultivated area in each farming region and dumping right allocation model under Trading System

In order to the technical scheme of the specific embodiment of the invention is described, first simply introduce ultimate principle of the present invention by example as follows.

Consider Interval Fuzzy two benches robust Multipurpose Optimal Method:

Objective function:

Maxλ ^±

$\mathrm{Min}0.2y_1^{\±}+0.8y_2^{\±}$

$\mathrm{Min}0.2v_1^{\±}+0.8v_2^{\±}$

Meet constraint condition:

$50x^{\±}0.2\×{6y}_1^{\±}-0.8\×8y_2^{\±}-0.2v_1^{\±}-0.8v_2^{\±}\≥50+{\mathrm{\λ}}^{\±}\×(100-50)$

$v_1^{\±}\≥6y_1^{\±}-0.2\×6y_1^{\±}-0.8\×8y_2^{\±}$

$v_2^{\±}\≥8y_2^{\±}-0.2\×6y_1^{\±}-0.8\×8y_2^{\±}$

$80x^{\±}-40y_1^{\±}\≤150+(1-{\mathrm{\λ}}^{\±})\×(200-150)$

$80x^{\±}-30y_2^{\±}\≤150+(1-{\mathrm{\λ}}^{\±})\×(200-150)$

x ^±≥0

$y_1^{\±}\≥0$

$y_2^{\±}\≥0$

$v_1^{\±}\≥0$

$v_2^{\±}\≥0$

λ ^±≥0

The first step: according to interactive remote teaching, write boundary's submodel:

Objective function:

Maxλ ⁺

$\mathrm{Min}0.2y_1^{-}+0.8y_2^{-}$

$\mathrm{Min}0.2v_1^{-}+0.8v_2^{-}$

Meet constraint condition:

$50x^{+}0.2\×6y_1^{-}-0.8\×8y_2^{-}-0.2v_1^{-}-0.8v_2^{-}\≥50+{\mathrm{\λ}}^{+}\×(100-50)$

$v_1^{-}\≥6y_1^{-}-0.2\×6y_1^{-}-0.8\×8y_2^{-}$

$v_2^{-}\≥8y_2^{-}-0.2\×6y_1^{-}-0.8\×8y_2^{-}$

$80x^{+}-40y_1^{-}\≤150+(1-{\mathrm{\λ}}^{+})\×(200-150)$

$80x^{+}-30y_2^{-}\≤150+(1-{\mathrm{\λ}}^{+})\×(200-150)$

x ⁺≥0

$y_1^{-}\≥0$

$y_2^{-}\≥0$

$v_1^{-}\≥0$

$v_2^{-}\≥0$

λ ^-≥0

Second step: according to interactive remote teaching, writes boundary's submodel:

Objective function:

Maxλ ^-

$\mathrm{Min}0.2y_1^{+}+0.8y_2^{+}$

$\mathrm{Min}0.2v_1^{-}+0.8v_2^{-}$

Meet constraint condition:

$50x^{-}0.2\×6y_1^{+}-0.8\×8y_2^{+}-0.2v_1^{+}-0.8v_2^{+}\≥50+{\mathrm{\λ}}^{-}\×(100-50)$

$v_1^{+}\≥6y_1^{+}-0.2\×6y_1^{+}-0.8\×8y_2^{+}$

$v_2^{+}\≥8y_2^{+}-0.2\×6y_1^{+}-0.8\×8y_2^{+}$

$80x^{-}-40y_1^{+}\≤150+(1-{\mathrm{\λ}}^{-})\×(200-150)$

$80x^{-}-30y_2^{+}\≤150+(1-{\mathrm{\λ}}^{-})\×(200-150)$

$0\≤x^{-}\≤x_{\mathrm{opt}}^{+}$

$y_1^{+}\≥y_{\mathrm{opt}}^{-}$

$y_2^{+}\≥y_{\mathrm{opt}}^{-}$

$v_1^{+}\≥v_{\mathrm{opt}}^{-}$

$v_2^{+}\≥v_{\mathrm{opt}}^{-}$

$0\≤{\mathrm{\λ}}^{-}\≤{\mathrm{\λ}}_{\mathrm{opt}}^{+}$

3rd step: solve bound submodel, finally can obtain optimum solution is ${\mathrm{\λ}}_{\mathrm{opt}}^{\±}=\left[\mathrm{0.525,1}\right];x_{\mathrm{opt}}^{\±}=2.09;y_{1\mathrm{opt}}^{\±}=0.4304;$ $y_{2\mathrm{opt}}^{\±}=0.5739;v_{1\mathrm{opt}}^{\±}=0;v_{2\mathrm{opt}}^{\±}=\left[\mathrm{0.4017,0.6313}\right]$

Known by above sample calculation analysis, Interval Fuzzy two benches robust Multipurpose Optimal Method can consider multiple target simultaneously, simultaneously between treatment region, Stochastic sum fuzzy uncertainty, obtains the optimum solution with unascertained information.

Based on above ultimate principle, now collect socioeconomic data, for model provides data basis, as shown in table 1-2.

Table 1 socioeconomic data

Blowdown flow rate data (unit 10 under the different emission level of table 2 ³kg)

Based on Interval Fuzzy two benches robust Multipurpose Optimal Method, to set up respectively under Trading System and not non-point source pollution control model under Trading System, not under Trading System, basin dumping right only considers primary distribution, Trading System down blow power is on original allocation basis, carry out the dumping right reallocation based on transaction, be shown below:

Under A, Trading System:

Objective function:

(1). system Maximum Satisfaction

Maxλ ^±

(2). total nitrogen total phosphorus discharge capacity is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E({\tilde{Y}}_{\mathrm{Nit}}+{\tilde{Y}}_{\mathrm{Pit}})$

(3). total nitrogen total phosphorus discharge capacity undulatory property is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}+{\widetilde{\mathrm{DE}}}_{\mathrm{Pit}})\right]$

Meet constraint condition:

$\begin{array}{c}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}{B}_{\mathrm{it}}^{\±}(S_{\mathrm{it}}^{-}+\mathrm{\α\Δ}{S}_{\mathrm{it}})-\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}+{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}})\right]\\ -\mathrm{\ρ}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}+{\widetilde{\mathrm{DE}}}_{\mathrm{Pit}})\right]\≥{\mathrm{\Ω}}_{\mathrm{opt}}^{-}+{\mathrm{\λ}}^{\±}({\mathrm{\Ω}}_{\mathrm{opt}}^{+}-{\mathrm{\Ω}}_{\mathrm{opt}}^{-})\end{array}$

$\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({\tilde{Q}}_N{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Nit}})\≤T_{\mathrm{Nt}}^{\±},\∀t$

$\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({\tilde{Q}}_P^{\±}{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Pit}})\≤T_{\mathrm{Pt}}^{\±},\∀t$

$0\≤S_{\mathrm{it}}^{\±}\≤S_{\mathrm{it}\max },\∀i,t$

${\tilde{Y}}_{\mathrm{Nit}}\≤{\tilde{Q}}_N{x}_{\mathrm{it}}^{\±},\∀i,t$

${\tilde{Y}}_{\mathrm{Pit}}\≤{\tilde{Q}}_P{x}_{\mathrm{it}}^{\±},\∀i,t$

${\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}\≥{\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}-E\left[{\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}\right],\∀i,t$

${\widetilde{\mathrm{DE}}}_{\mathrm{Pit}}\≥{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}}-E\left[{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}}\right],\∀i,t$

0≤λ ^±≤1

Under B, nontransaction system:

Objective function:

(1). system Maximum Satisfaction

Maxλ ^±

(2). total nitrogen total phosphorus discharge capacity is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E({\tilde{Y}}_{\mathrm{Nit}}+{\tilde{Y}}_{\mathrm{Pit}})$

(3). total nitrogen total phosphorus discharge capacity undulatory property is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}+{\widetilde{\mathrm{DE}}}_{\mathrm{Pit}})\right]$

Meet constraint condition:

$\begin{array}{c}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}{B}_{\mathrm{it}}^{\±}(S_{\mathrm{it}}^{-}+\mathrm{\α\Δ}{S}_{\mathrm{it}})-\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}+{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}})\right]\\ -\mathrm{\ρ}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E\left[({\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}+{\widetilde{\mathrm{DE}}}_{\mathrm{Pit}})\right]\≥{\mathrm{\Ω}}_{\mathrm{opt}}^{-}+{\mathrm{\λ}}^{\±}({\mathrm{\Ω}}_{\mathrm{opt}}^{+}-{\mathrm{\Ω}}_{\mathrm{opt}}^{-})\end{array}$

${\tilde{Q}}_N{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Nit}}\≤(\mathrm{\α}\frac{P{Q}_i}{\underset{i}{\mathrm{\Σ}}P{Q}_i}+\mathrm{\β}\frac{{\mathrm{GDP}}_i}{\underset{i}{\mathrm{\Σ}}{\mathrm{GDP}}_i}+\mathrm{\χ}\frac{Q_i}{\underset{i}{\mathrm{\Σ}}{Q}_i})T_{\mathrm{Nt}}^{\±},\∀t$

${\tilde{Q}}_P{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Pit}}\≤(\mathrm{\α}\frac{P{Q}_i}{\underset{i}{\mathrm{\Σ}}P{Q}_i}+\mathrm{\β}\frac{{\mathrm{GDP}}_i}{\underset{i}{\mathrm{\Σ}}{\mathrm{GDP}}_i}+\mathrm{\χ}\frac{Q_i}{\underset{i}{\mathrm{\Σ}}{Q}_i})T_{\mathrm{Pt}}^{\±},\∀t$

$0\≤S_{\mathrm{it}}^{\±}\≤S_{\mathrm{it}\max },\∀i,t$

${\tilde{Y}}_{\mathrm{Nit}}\≤{\tilde{Q}}_N{S}_{\mathrm{it}}^{\±},\∀i,t$

${\tilde{Y}}_{\mathrm{Pit}}\≤{\tilde{Q}}_P{x}_{\mathrm{it}}^{\±},\∀i,t$

${\widetilde{\mathrm{DE}}}_{\mathrm{Nit}}\≥{\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}-E\left[{\mathrm{EP}}_{\mathrm{Ni}}^{\±}{\tilde{Y}}_{\mathrm{Nit}}\right],\∀i,t$

${\widetilde{\mathrm{DE}}}_{\mathrm{Pit}}\≥{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}}-E\left[{\mathrm{EP}}_{\mathrm{Pi}}^{\±}{\tilde{Y}}_{\mathrm{Pit}}\right],\∀i,t$

0≤λ ^±≤1

Wherein, the meaning of parameters that represents of each symbol is as follows:

+ and-represent the upper and lower bound of interval parameter respectively;

~ represent non-linear stochastic parameter;

I refers to farming region; T is project period; J is probability level;

Ω _optit is the aspiration level of system net proceeds;

λ is decision maker about the comprehensive satisfaction of system benefit and environmental pollution;

B _itthe unit area land revenue (RMB$) of i farming region when t project period;

S _itthe land area (ha) of i farming region when t project period;

π is decision-making coefficient;

ρ is the risk level under kinds of risks sight;

EP _niand EP _pithe fine (RMB$) exceeded standard suffered by 1kg blowdown respectively;

Y _nitand Y _pittotal nitrogen and total phosphorus discharge beyond standards amount (kg) respectively;

DE _nijtand DE _pijtthat i region discharge beyond standards total nitrogen and total phosphorus receive environment fine and the deviation of imposing a fine average level respectively

(RMB￥)；

Q _nijand Q _pijthe unit area blowdown flow rate (kg) of i farming region under j probability level respectively;

T _ntand T _ptthe total emission volumn standard (kg) about total nitrogen total phosphorus in basin respectively;

PO _iit is the size of population (people) of i farming region;

GDP _iit is the total output value of i farming region;

α, β, χ are the weight of population, total output value, blowdown flow rate in dumping right assigning process respectively.

Four, model solution step

1. systematic collection and the fragrant small stream river valley data of arrangement, set up fragrant small stream river valley hydrological model, described data comprise hydrographic data, watershed unit data, river network of watershed data, Soil attribute data, weather data for many years, land use data etc., also comprise the agricultural planting cost of chief crop type in basin, annual production, agricultural product price, cultivated area and agricultural chemicals, chemical fertilizer, charges for water and electricity.

2. pair hydrological model set up carries out calibration and checking, based on the hydrological model of checking, carries out hydrologic(al) prognosis modeling effort.

3. based on base flow separation technology, from channel flow process, extract rainwash data, obtain overland flow hydrograph project period.

4., based on Monte Carlo technique matching rainwash probability cumulative distribution function, pair distribution function carries out sliding-model control, obtains the rainwash discrete data under multiple probability level.

5., based on rainwash discrete data estimation non-point source load, and by result of calculation unbalanced input non-point source pollution control plan model, realize docking of analogue technique and Optimized model.

6. pair non-linear non-point source pollution control plan model adopts interactive remote teaching, splits submodel, carries out linearization respectively solve upper bound submodel and lower bound submodel.

7., by changing risk level ρ, define 11 kinds of risky situations, under 11 kinds of risky situations, risk level value is respectively 0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.Risk level represents the attention degree of supvr to system risk, risk level is higher, show that supvr more payes attention to system risk, such as, ρ=1 shows that supvr considers the impact of system risk on system in earnest, by system risk value based on the thought of risk averse in non-point source Emission trading planning process incorporate in decision process; ρ=0 shows that supvr gives no thought to system risk based on the thought of risk-neutral in non-point source Emission trading planning process, planning risk value be defaulted as 0; ρ=0.5 shows that supvr considers system risk in non-point source Emission trading process, and by the real system value-at-risk of 0.5 times, namely include in the calculating of decision process.

8., by solving linear equation, obtain agriculture optimum cultivated area under these 11 kinds of risky situations and each farming region is by handing over facile optimum dumping right and the comprehensive satisfaction of environmental pollution under obtaining 11 kinds of risky situations

At laboratory calculating simulation solving model, under obtaining Trading System and the not optimum cultivated area in each farming region under Trading System, as Fig. 3 transaction with not under mechanism of exchange shown in optimum cultivated area.Cultivated area is more, more agricultural income can be brought, but, also can bring more serious non-point pollution problem, on the contrary, cultivated area is fewer, and agricultural income can reduce, but, environmental risk also can reduce, cultivated area under Trading System is than few under not Trading System, and illustrate that model system is optimized whole farming region cultivated area under Trading System, the introducing of Trading System and the foundation of optimum Emission trading mechanism significantly can cut down the discharge capacity of nutriment.Within project period, altogether decrease the total nitrogen of [213.7,288.8] ton and the discharge of [11.8,30.2] ton total phosphorus.Although cultivated area decreases, under Trading System, income adds [6.6,9.5] × 10 ⁶yuan; through quantitative test; absolutely prove and non-linear agricultural nonpoint source pollution controlling planning model system is applied in the foundation of optimum Emission trading mechanism; when can ensure that income increases; the discharge of remarkable attenuating Non-point Source Pollutants; while sustaining economic growth, protect environment again, really accomplish agriculture sustainable development in fragrant small stream river valley demonstration area.

Five, optimum Emission trading mechanism is set up

Reference point source emission standard and each farming region level of economic development, reasonable assumption fragrant small stream river valley each farming region emission standard value, by the comprehensive computing of model system, obtain the dumping right of each farming region reallocation under agriculture Emission trading mechanism, the difference of getting both obtains the dealing situation that each farming region is concrete under transaction framework, as shown in Fig. 4 total nitrogen Emission trading best mechanism, on the occasion of representing the amount sold, negative value then represents the dumping right amount bought in.Because the discharge that exceeds the quata can cause huge environment fine, the farming region of super row is considered for interests, be more prone to spend less money to buy in more dumping right, again because each district agricultural net benefits has different, blowdown flow rate is different again, and thus optimum trading volume is also different, by the mechanism of this optimization distribution environments resource, avoid a large amount of super row's fine, optimize system benefit.

Claims (6)

1. a non-linear agricultural nonpoint source pollution control method, is characterized in that, comprise the steps:

S1. collect and arrange basin natural data and social data, in conjunction with two stage stochastic programming method, Interval Programming method, random robust planing method and system ambiguous planing method, build Interval Fuzzy two benches robust Model for Multi-Objective Optimization and Watershed Hydrologic Models, and water basin combination literary composition model, simulation flow anomaly physical mechanism, prediction rainwash;

S2. set up hydrological model is verified, based on the hydrological model after checking, carry out hydrologic(al) prognosis modeling effort;

S3. based on base flow separation technology, from channel flow process, extract rainwash data, obtain overland flow hydrograph;

S4. based on Monte Carlo technique matching rainwash probability cumulative distribution function, pair distribution function carries out sliding-model control, obtains the rainwash discrete data under multiple probability level, sets up objective function and constraint condition, builds and solving-optimizing model;

S5. based on rainwash discrete data estimation non-point source load, by result of calculation unbalanced input non-point source pollution control model, docking of analogue technique and Optimized model is realized;

S6. interactive remote teaching is adopted to non-linear non-point source pollution control model, split submodel, respectively linearization is carried out to upper bound submodel and lower bound submodel and solve;

S7. by changing risk level ρ, define 11 kinds of risky situations, under 11 kinds of risky situations, risk level value is respectively 0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1;

S8. by solving linear equation, agriculture optimum cultivated area under these 11 kinds of risky situations is obtained and each farming region is by handing over facile optimum dumping right and the comprehensive satisfaction of environmental pollution under obtaining 11 kinds of risky situations

2. method according to claim 1, it is characterized in that, natural data described in step S1 and social data comprise watershed unit data, river network of watershed data, Soil attribute data, for many years weather data, agrotype, annual production, agricultural product price, cultivated area and comprise the agricultural planting cost of agricultural chemicals, chemical fertilizer, charges for water and electricity.

3. method according to claim 1, is characterized in that, Watershed Hydrologic Models described in step S1 is Watershed-scale distributed hydrological model, and the algorithm of employing is:

${\mathrm{SW}}_t={\mathrm{SW}}_0+\underset{i=1}{\overset{t}{\mathrm{\Σ}}}(R_{\mathrm{day}}-Q_{\mathrm{surf}}-E_a-w_{\mathrm{seep}}-Q_{\mathrm{gw}})$

$Q_{\mathrm{surf}}=\frac{{(R_{\mathrm{day}}-I_a)}^2}{(R_{\mathrm{day}}-I_a+S)}$

$S=25.4(\frac{1000}{\mathrm{CN}}-10)$

Wherein, the meaning of parameters that represents of each symbol is as follows:

SW _tsoil moisture content (mm H ₂o);

SW ₀initial soil moisture content (mm H ₂o);

R _daydaily rainfall (mm H ₂o);

Q _surfa day flow path surface (mm H ₂o);

E _adaily evaporation amount (mm H ₂o);

W _seep(the mm H of milliosmolarity under the holard ₂o);

Q _gwunderground water diurnal courses amount (mm H ₂o);

I _arainfall spurt value (mm H ₂o);

S is hysteresis index;

CN is runoff curve number;

I is time (day);

T is duration (day).

4. method according to claim 1, is characterized in that, described step S4 builds and in solving-optimizing model; By systematic uncertainty with the form body of random number and interval number now in this process.

5. method according to claim 1, is characterized in that, to set up respectively under Trading System and not non-point source pollution control model under Trading System in described step S7, as follows:

Under A, Trading System:

Objective function:

(1). system Maximum Satisfaction

Maxλ ^±

(2). total nitrogen total phosphorus discharge capacity is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E({\tilde{Y}}_{\mathrm{Nit}}+{\tilde{Y}}_{\mathrm{Pit}})$

(3). total nitrogen total phosphorus discharge capacity undulatory property is minimum

Meet constraint condition:

$\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({\tilde{Q}}_N{S}_{\mathrm{ii}}^{\±}-{\tilde{Y}}_{\mathrm{Nit}})\≤T_{\mathrm{Nt}}^{\±},\∀t$

$\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({{\tilde{Q}}_P^{\±}S}_{\mathrm{ii}}^{\±}-{\tilde{Y}}_{\mathrm{Pit}})\≤T_{\mathrm{Pt}}^{\±},\∀t$

$0\≤S_{\mathrm{it}}^{\±}\≤S_{\mathrm{it}\max },\∀i,t$

${\tilde{Y}}_{\mathrm{Nit}}\≤{\tilde{Q}}_N{x}_{\mathrm{it}}^{\±},\∀i,t$

${\tilde{Y}}_{\mathrm{Pit}}\≤{\tilde{Q}}_P{x}_{\mathrm{it}}^{\±},\∀i,t$

0≤λ ^±≤1

Under B, nontransaction system:

Objective function:

(1). system Maximum Satisfaction

Maxλ ^±

(2). total nitrogen total phosphorus discharge capacity is minimum

$\mathrm{Min}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{t=1}{\overset{k}{\mathrm{\Σ}}}E({\tilde{Y}}_{\mathrm{Nit}}+{\tilde{Y}}_{\mathrm{Pit}})$

(3). total nitrogen total phosphorus discharge capacity undulatory property is minimum

Meet constraint condition:

${\tilde{Q}}_N{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Nit}}\≤(\mathrm{\α}\frac{{\mathrm{PO}}_i}{\underset{i}{\mathrm{\Σ}}{\mathrm{PO}}_i}+\mathrm{\β}\frac{{\mathrm{GDP}}_i}{\underset{i}{\mathrm{\Σ}}{\mathrm{GDP}}_i}+\mathrm{\χ}\frac{Q_i}{\underset{i}{\mathrm{\Σ}}{Q}_i})T_{\mathrm{Nt}}^{\±},\∀t$

${\tilde{Q}}_P{S}_{\mathrm{it}}^{\±}-{\tilde{Y}}_{\mathrm{Pit}}\≤(\mathrm{\α}\frac{{\mathrm{PO}}_i}{\underset{i}{\mathrm{\Σ}}{\mathrm{PO}}_i}+\mathrm{\β}\frac{{\mathrm{GDP}}_i}{\underset{i}{\mathrm{\Σ}}{\mathrm{GDP}}_i}+\mathrm{\χ}\frac{Q_i}{\underset{i}{\mathrm{\Σ}}{Q}_i})T_{\mathrm{Pt}}^{\±},\∀t$

$0\≤S_{\mathrm{it}}^{\±}\≤S_{\mathrm{it}\max },\∀i,t$

${\tilde{Y}}_{\mathrm{Nit}}\≤{\tilde{Q}}_N{x}_{\mathrm{it}}^{\±},\∀i,t$

${\tilde{Y}}_{\mathrm{Pit}}\≤{\tilde{Q}}_P{x}_{\mathrm{it}}^{\±},\∀i,t$

0≤λ ^±≤1

Wherein, the meaning of parameters that represents of each symbol is as follows:

+ and-represent the upper and lower bound of interval parameter respectively;

~ represent non-linear stochastic parameter;

I refers to farming region; T is project period; J is probability level;

Ω _optit is the aspiration level of system net proceeds;

λ is decision maker about the comprehensive satisfaction of system benefit and environmental pollution;

B _itthe unit area land revenue (RMB$) of i farming region when t project period;

S _itthe land area (ha) of i farming region when t project period;

π is decision-making coefficient;

ρ is the risk level under kinds of risks sight;

EP _niand EP _pithe fine (RMB$) exceeded standard suffered by 1kg blowdown respectively;

Y _nitand Y _pittotal nitrogen and total phosphorus discharge beyond standards amount (kg) respectively;

DE _nijtand DE _pijtthat i region discharge beyond standards total nitrogen and total phosphorus receive environment fine and the deviation (RMB$) of imposing a fine average level respectively;

Q _nijand Q _pijthe unit area blowdown flow rate (kg) of i farming region under j probability level respectively;

T _ntand T _ptthe total emission volumn standard (kg) about total nitrogen total phosphorus in basin respectively;

PO _iit is the size of population (people) of i farming region;

GDP _iit is the total output value of i farming region;

α, β, χ are the weight of population, total output value, blowdown flow rate in dumping right assigning process respectively.

6. method according to claim 1, it is characterized in that, said optimum dumping right in described step S8, its mechanism of exchange is for passing through model calculation, obtain the dumping right of each farming region reallocation under agriculture Emission trading mechanism, the difference of getting both obtains each farming region under transaction framework and specifically buys and sells situation.