Figure 2 Calculation interface
in a [selected] way. For example, you can debug the [flow only] model by selecting only the [flow constraints] group. You can debug the [flow only + specific gravity] model by selecting the [flow constraints] group and the [specific grav - ity] constraint. If the [DB structure] parameter is not selected, the DB calculation will not be performed, and so on. • Nonlinear programming (NLP) global convergence strat- egy: GROMS strategically achieves the convergent optimal of nonlinear models and has a high probability of obtaining the global optimal solution. There could always be a local optimum; if it occurs, it can be eliminated by adjusting the calculation parameters and recalculating. In testing 300 model cases from 30 refining and chemical enterprises, it was found that the local optimal solution probability is approximately 5%. This is far lower than the 30% proba- bility of a local optimal solution of other programs. Other programs also struggle or fail to converge on a complex or large enterprise model used in the GROMS testing. • Determination of the baseline value of global optimal solution for nonlinear problems: Production planning models for a refinery are one-dimensional linear models
where physical properties are not considered (flow only). A unique global optimal solution can be obtained by solv - ing this model. A nonlinear production planning model is a one-dimensional linear model that adds the constraints of physical properties. The nonlinear model is a deriva - tive model of the one-dimensional linear model. The less constrained one-dimensional model’s objective function is greater than that of the nonlinear production model. A series of tests using GROMS and another commer- cially available linear programming (LP) program were conducted. The GROMS one-dimensional model objec- tive function is the baseline performance against which to measure its nonlinear model and the other program. Each model was perturbed by increasing the gasoil hydrocrack- ing unit yields of light diesel with an offsetting decrease of residue. Increments of 1% yields were individual cases. The resulting objective function values are plotted in Figure 4 and reported in Table 1 . GROMS reached the global optimum while the other LP program has a local optimum below GROMS and the global optimum.
Overall economic benefits of enterprises in the region
Items
Total
* Qing
* Jiang
* Han
* Men
* Nan
Crude oil processing capacity (10,000 tons) Sales revenue (10,000 yuan) Raw material cost (10,000 yuan) Variable cost (10,000 yuan) Gross profit (10,000 yuan) Gross profit per ton (yuan/ton)
292
58
58
62
45
69
1,656,167 1,577,335
339,729 322,601
356,650 331,434
334,596 323,793
248,201 240,999
376,990 358,507
34,193 44,639
7,463 9,666
5,601
8,274 2,529
4,982 2,220
7,873
19,615
10,610
140
148
293
39
46
147
Table 1
89
PTQ Q2 2025
www.digitalrefining.com
Powered by FlippingBook